WitMotion WT901C-TTL mpu9250 High-Precision Gyro+Accelerometer+Angle+Magnetometer,9-Axis Digital Compass

[Pablo Lewin]

https://www.amazon.com/gp/product/B01N03WKDV/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8&psc=1

I trying to implement an Alt/Az display for my home made motorized mount, so I purchased on Amazon the WitMotion WT901C-TTL mpu9250 High-Precision Gyro+Accelerometer+Angle+Magnetometer,9-Axis Digital Compass 

For 34 dollars...

It comes with a pretty interesting and seemingly useful software package and it's easy to install.

I will report with my results soon, in the meantime does anybody have any experience with this device?

Pablo Lewin WA6RSV

[Hamish Barker]

no experience with that one, but I have had good experience with a hybrid system using an encoder for azimuth and 6dof gyro/mag chip for altitude. See the thread on cloudy nights.com here:

https://www.cloudynights.com/topic/589521-37-dobsonian-dsc-for-diy-makers/#entry8069312

there are a couple of different threads. one with the hybrid, and another with two encoders. Gyro chip etc doesn't give terribly accurate (circa 0.5 degree) altitude. More than accurate enough for radio though. just bearable for optical telescopes (0.25 degrees is better to ensure objects appear in the eyepiece reliably).

the encoders used are quite cheap, about $10-15 on aliexpress because they are either 600 or 1200 steps per rotation, but by gearing up with ubiquitous 2mm pitch timing belts (also cheap at aliexpress or other maker supply stores) achieve 2400-4800 steps per rotation. Typical digital setting circles for optical telescopes use single turn encoders with circa 10,000 steps/turn, but those encoders are rather more expensive, circa $100 each.

the processor is an esp32 with built in wifi, programmed via arduino development environment or visual studio as you prefer. the code (see the github link via the cloudynights thread) acts as a wifi point which serves encoder count (alt and az) back to planetarium software such as sky safari by a very basic html query. (just provides two numbers back on query, so would be easy to program interface to your own software if skysafari is not suitable.

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/3190ec43-64f9-460f-af56-526f7c21640fn%40googlegroups.com.

[Larry Mayfield]

Just off hand this looks interesting!  There are some issues, maybe like this one

 

“Note: Z-axis angle will be affected by magnetic field. It is required to calibrate and keep the sensor at least 20cm away from the stuff like metal that may create magnetic field interference

 

20 centimeters isn’t much, about 8 inches or so…I would mount one on the dish pointing axis.

 

Banggood has some 9 axis gyro units so maybe I’ll take a look at what they have although this price seems pretty fair as it includes software as well. Just for chuckles, I have a timex expedition watch (circa, 2010)  that calibrates its compass the same way…

 

Larry

Pahrump, NV

--

[mike....@gmail.com]

First thoughts were:

1) how far is the antenna from the control station in the observatory?  The output of the one linked above was ttl serial at 3 or 5 volts. This would be good to 10-15 ft? old rs-232 said never to go over 50 ft (12v level).  Could be increased using RS-422 translators(thousands of feet). Or rf links like HC-12 or wifi but then you are generating RF. So this isn't unique to this sensor because this will be need to be  solved for what ever sensor.

2) software gives a lot of readouts but are any of them directly usable to the operator?  in astronomy usually you work in  "AZEL"  or "HA/DEC".  May need a translational program to calculation your antenna orientation in astro units?

3) What are the temperature specs?  I scan through the specs but did not see them.

I am not knocking this down. It looks useful.   We'll be waiting for your evaluation.

mike

[fasleitung3]

Pablo,

we have been investigating MEMS devices including the MPU9250 for their usefulnes for our radio telescopes. The finding is that they are sufficient measuring the elevation of smaller telescopes. We actually use them on our 1.2-m and 2.3-m telescopes and on our Ku-Band interferometer for that purpose. What we did find was that the magnetic compass of the MPU9250 was not sufficient for long term accurate positioning in azimuth. The main reasons were long term stability and susceptibiltiy to external magnetic fields.

We started using MEMS some time ago and at that time the ADXL 203 was available. So all our devices in the field are based on this unit. For newer installations we will probably consider the MPU9250 or it's successor, the ICM-20948 as the MPU9250 did perform well in the lab (but only for elevation).

We combine the sensor with an Arduino which takes care of some processing and the communication to the telescope control system. This is enclosed in a watertight housing as shown in the picture attached. This is from our 2.3-m telescope.

Good luck with your experiments,

Wolfgang

--

[Pablo Lewin]

Thank you for your input. What software did you use to readout the encoder? I see no mention in the CN post or pictures of the software.

[Pablo Lewin]

Larry my mounting strategy is to attach it to a 1 meter long piece of wood connected to the dish. The unit is extremely small and feather light weight. All I need is to come to withing 5 degrees of my heading for it to be useful. We'll see. I'm having trouble making it work with my Radio Telescope computer because the driver is clashing with the LNA 50 ohm switch usb driver... lots of wor

[Pablo Lewin]

Hi Mike

The computer system (control station)  is at the bottom of the antenna,it's a full out mini Windows 10 PC with NO WIFI , only an ethernet connection going to it. It has a monitor dummy and I'm running it headless connecting through Splashtot.

I Installed the sensor on one end of a 1 meter piece of wood...The system works well on my main computer but so far my Radio Telescope computer can't detect it because it seems to be clashing with another driver so I will have to find a solution for that.

The software has an airplane type compass display and an "attidtide indicator" for elevation and bank (tilt).

I don't know what the temp specs are...I still have to make it work with my computer I'll worry about it later

Pablo

[Pablo Lewin]

Thank you Wolfgang 

I see that your devices are attached directly to metal and that MAY have a bearing on the Azimuth problem. The strategy I'm using is to attach the sensor to a 1 meter wooden pole /rafter (One could use a PVC pipe) anything  to keep the sensor at least a meter away from the antenna while providing minimum flexure, thus avoiding the potential magnetic interference with the sensor from metal.  I stole the concept from the old MD-80's I used to fly for a living whereby the compass "flux gate" was away from the cockpit and near the right wingtip (Not to be confused with the simple Compass which was located behind the Co-pilot's seat but I digress).

Pablo

[fasleitung3]

Hi Pablo,

The device shown in the picture does not even have a compass. I am referring to some tests which we did both in the lab and by trying to use it on a telescope. Both were not satisfactory, but your experience may be different as your setup may be more benign with respect to magnetic influence.

Wolfgang

[Paul Oxley]

Pablo

I did some experiments with a similar setup. I even tried mounting the compass sensor on a non-ferrous (aluminum) angle piece. It didn't help.

Seems to me that the Accelerometer solution would depend on having motion present for the measurement. With the antenna motion, it likely isn't fast enough to sense this way. The acceleration motion would be in bursts when the motion started and ended.

An inclinometer sensor did work. It depends on gravity for the measurement.

Paul

https://groups.google.com/d/msgid/sara-list/5de3902f-4bcf-4e7f-8f65-0589cd322559n%40googlegroups.com

.

[Pablo Lewin]

Hi Paul and everyone, I flimsily attached the sensor to my scope and as predicted the elevation is spot on, but the "heading" is way off..but wait, although the heading is way off it is actually stable and it turns and I still haven't gone through the calibration scheme which I will try tomorrow. It's a primitive system but for the money I will work with it until I either get it right or..not.

The Magnetic variation in the Los Angeles area is -13 degrees (East) so I have to keep that in mind when I choose a heading from Radio Eyes.

Pablo

[Raydel Abreu (CM2ESP)]

Hi all

Magnetometer sensors for heading are quite hard to calibrate and a little bit tricky to calculate under some circumstances.

My project here doesn't have a very accurate sensor. It is a GY-86 IMU board.

The problem with headings via magnetometer that I am having (and probably you too) is that you need to run a very accurate calibration in all 3 axis. So you can make an estimation of minimum, maximum, axis and even multiplication factor. Ferromagnetic elements in the vicinity even couple of meters introduces errors. In order to get an accurate heading you need to do the maths based in an hypothetical sphere (3 axis) or a perfect circle (2 axis). But nearby objects may affect one axis more than other and what should look as a circle (X axis min/max and Y axis min/max centered in 0;0) then looks like an ellipse which might be also slightly shifted in one axis. So an accurate min/max/offset/multiplication factor is needed so you can more or less turn that ellipse into something more "circular".

If you are calculating heading in a horizontal system you just need to do this calibration for axis X and Y. But this doesn't work if your antenna is tilted. Then you need to add Z axis to the problem. However.... calibrating a bare sensor in 3 axis is easy, just move all 3 axis in full rotation. But once you have your sensor embedded in your antenna. Then things becomes messy. How can you do a 360 degree movement on the "Elevation" axis in order to "calibrate" Z axis if you have a big dish on it???

I am no expert at all and thia problem (calibrating Z axis) is something I haven't solve yet with rudimentary skills. The problem is that while you can do the calibration in 3 axis with the bare sensor, that calibration will then be useless once you have added all the rest of the antenna hardware, unless the 100% of the components were not ferromagnetic, which is almost impossible with all the connectors, cables, LNA casing, etc.

With my little to barely none knowledge this problem is still pending. What I have done as a temporary solution is to do a calibration routine for the antenna when power-up. On this calibration the antenna is pointed at the start to Elevation 90 (pointing zenith) so the magnetometer is then perfect horizontally. Then the antenna rotates left/right couple of times so calibration factors on X/Y are calculated. When calibration is ready, the system is ready for Transit observation (not tracking an object/source). If I want to go to an specific place then the antenna first goes straight up to 90 elevation, then azimuth is rotated and an accurate heading is obtained, finally antenna is moved to the needed elevation, and due to effect of Z axis the reported heading is wrong, but actually it is at the right spot. This method is useless for tracking but for transit observations it is a simple dirty solution of a Go/To dish.

Of course, there are math formulas to avoid this and compensate the effect of Z axis on a tilted magnetometer, but again, how you rotate 360 degrees or even 180 degrees dish/telescope in the elevation axis?

Sorry for the long post with little useful info. I just wanted to share my tales when dealing with magnetometer for azimuth.

If anyone want more info or details about my rotor just drop me a mail.

Regards.

Raydel, CM2ESP

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/d06d6f2a-bdc4-424d-a757-9754dce8b9bdn%40googlegroups.com.

[Larry Mayfield]

Paul, did you come up with a good answer for the sensors? I agree with you that mag field sensors are suspect and the earths field changes a lot depending on solar winds etc. I still like rotary encoders for my project.  Accelerometer might work if a triaxial unit is used.   Signal conditioning would be a key element for any of them though.   I think any that I use would be a stepper or servo  type motor with an encoder on the end of  the shaft to count degrees or pulses and then back calculate for degrees  (for data recording) and actual position. And of all things, after seeing the example of the sensor being used on  the  Celestron optical unit, I realized that  I had been thinking  inside the box, lol..  I have been trying to measure the  movement of the final output when I can measure  the input which is easier since it urns a lot more than microscopically…

 

An inclinometer  could even work for an az measuring set up if a gear driven unit with a bevel gear   Just need units, with digital or analog output for signals, is used.

 

Back to my second cuppa now…

 

And thanks for reminding me of the use of inclinometers..

 

Larry

Pahrump, NV

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/1243499776.19466.1612826747263%40mail.yahoo.com.

[Larry Mayfield]

Paul, all,  number 2 reply, lol, I am old and keep forgetting stuff..

 

Here is the Witmotion device I would use. This is a link to an amazon page

 

WitMotion WT61C TTL 3 Axis 3-Axis 6 Axis 6-Axis Digital Accelerometer Gyro Gyroscope inclinometer UART TTL MPU6050 MPU-6050 Module Triaxial Angle Acceleration Output: Amazon.com: Industrial & Scientific

 

With sensors of this type they have a mass on the active side and the change if load due to gravity is what is measured (I think).  Also scroll down this page along ways and  you can find sample  data and code locations where you  click  and find stuff. I down loaded a pile of them yesterday;  included as attachments for this one is the data sheet and user’s manual which identifies sample code and where to get it. It is available for many microcontrollers and SBC units.  (I hope I picked the correct pdf files to include, lol). And  I did download the code etc.

 

Larry

Pahrump, NV

 

 

 

From: sara...@googlegroups.com [mailto:sara...@googlegroups.com] On Behalf Of Paul Oxley

Sent: Monday, February 8, 2021 3:26 PM

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/1243499776.19466.1612826747263%40mail.yahoo.com.

[Paul Oxley]

Larry

The mean of earth's gravity points to the center of the earth. This is what is used to sense the elevation angle.

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/000001d6feed%24b84719e0%2428d54da0%24%40mayfco.com.

[Larry Mayfield]

But of course, lol. However, accelerometers measure  the “normal”  or restoring force. And that  is through the centerline of the sensor and with alignment to the earth’s center as you say. But if you take that sensor and move it off angle (ie tilt it, It will change its reading  because the Normal force is now  different due to the angle of it is making.     If solidly mounted to the axis of rotation will have all three axis producing a varying amount of new normal  or restoring forces.  For an azimuth drive system to measure the rotation angle, then one of the axes would have to be placed on  something that rotates in accordance with the AZ drive so that a variance if normal force could be determined.  I mentioned that  a bevel gear arrangement with say the x axis accel horizontal and the axle rotated to change the angle of the accelerometer can be used to measure that angular rotation. Of course it would need calibration and signal conditioning, but all doable.  The altitude angular measurement could use a simple one axis accelerometer with output. It is, on this kind of mount, just basically a rotator system.  I will look through my  pile of arduino modules and see if there is an accelerometer and if so see if I can gin up a test  rig to demonstrate or show my ignorance, either is possible and likely,  lol.

 

Oh, here is a snippet from the web…

 

Force Balance Sensor

Force Balance Sensor

These sensors are gravity referenced sensors and are anticipated for DC acceleration measurements like ships, vehicles, aircraft and seismic events. These sensors are frequently used in inclinometers and tilt meters. Force balance sensors are capable of measuring levels from 0.0001g to 200g, and the frequency range is from DC to 1000Hz. The advantages of these sensors include their high accuracy, a change in broad measurement, insensitivity to temperature change and their high accuracy.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/1040829168.329009.1612891122075%40mail.yahoo.com.

[Paul Oxley]

Larry

When the antenna is tilted with an inclinometer attached, the reference of the inclinometer changes. The gravity sensor still points at the center of the mass of the earth. It is measuring the change in the reference system.

I doubt that an acceleration based system could measure the change in the azimuth of the antenna. The antenna is stationary except when the change occurs with its movement to a different azimuth. Then it is stationary again.

I have a wild idea that I intend to pursue. I wonder if one could measure the difference in lat/long using two GPS units attached at a fixed distance (i.e. 1 M on a non-ferrous material). Both of the GPS units should see the same satellite due to their short distance apart. This should cancel out some of the errors in the GPS measurement.

The GPS measurements would need to be converted to an X & Y position  (in Meters) should be doable. Using simple geometry, the angle (referenced to the earth) of the separating support could be calculated.

I have two small GPS units that were intended for use on the roof of a car. They attach magnetically to the roof. They have USB interfaces which would be used with a computer to do the calculations. The units are labeled "GNSS Receiver" from QGP supply. They do not show any model number. I think I bought them at a local auto supply store.

Does anyone have experience with GNSS (GPS) units mounted on close distances? Is 1M sufficient spacing to obtain a differential GPS measurement accurately?

If this method works, it would be necessary to attach the separating support to a dish antenna retaining sufficient exposure to receive the GPS signal which comes from GPS satellites in polar orbits.

What do you think?

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/003e01d6ff1c%24f3b08f50%24db11adf0%24%40mayfco.com.

[Captain Anne Flint]

Hi Paul, 

I used to work with commercial grade GPS units; if it were me, I’d look at the unit/error specs to see if they are accurate to within one meter (6 meters just floated up in my brain, but that was a really long time ago) and move them out past their error rating[s]. 

Wende 

On Feb 9, 2021, at 3:26 PM, Paul Oxley <oxl...@att.net> wrote:

Larry

When the antenna is tilted with an inclinometer attached, the reference of the inclinometer changes. The gravity sensor still points at the center of the mass of the earth. It is measuring the change in the reference system.

I doubt that an acceleration based system could measure the change in the azimuth of the antenna. The antenna is stationary except when the change occurs with its movement to a different azimuth. Then it is stationary again.

I have a wild idea that I intend to pursue. I wonder if one could measure the difference in lat/long using two GPS units attached at a fixed distance (i.e. 1 M on a non-ferrous material). Both of the GPS units should see the same satellite due to their short distance apart. This should cancel out some of the errors in the GPS measurement.

The GPS measurements would need to be converted to an X & Y position  (in Meters) should be doable. Using simple geometry, the angle (referenced to the earth) of the separating support could be calculated.

I have two small GPS units that were intended for use on the roof of a car. They attach magnetically to the roof. They have USB interfaces which would be used with a computer to do the calculations. The units are labeled "GNSS Receiver" from QGP supply. They do not show any model number. I think I bought them at a local auto supply store.

Does anyone have experience with GNSS (GPS) units mounted on close distances? Is 1M sufficient spacing to obtain a differential GPS measurement accurately?

If this method works, it would be necessary to attach the separating support to a dish antenna retaining sufficient exposure to receive the GPS signal which comes from GPS satellites in polar orbits.

What do you think?

Paul

On Tuesday, February 9, 2021, 02:51:37 PM EST, Larry Mayfield <drm...@mayfco.com> wrote:

But of course, lol. However, accelerometers measure  the “normal”  or restoring force. And that  is through the centerline of the sensor and with alignment to the earth’s center as you say. But if you take that sensor and move it off angle (ie tilt it, It will change its reading  because the Normal force is now  different due to the angle of it is making.     If solidly mounted to the axis of rotation will have all three axis producing a varying amount of new normal  or restoring forces.  For an azimuth drive system to measure the rotation angle, then one of the axes would have to be placed on  something that rotates in accordance with the AZ drive so that a variance if normal force could be determined.  I mentioned that  a bevel gear arrangement with say the x axis accel horizontal and the axle rotated to change the angle of the accelerometer can be used to measure that angular rotation. Of course it would need calibration and signal conditioning, but all doable.  The altitude angular measurement could use a simple one axis accelerometer with output. It is, on this kind of mount, just basically a rotator system.  I will look through my  pile of arduino modules and see if there is an accelerometer and if so see if I can gin up a test  rig to demonstrate or show my ignorance, either is possible and likely,  lol.

 

Oh, here is a snippet from the web…

 

Force Balance Sensor

<image002.jpg>Force Balance Sensor

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/1910722943.448359.1612905985528%40mail.yahoo.com.
<image002.jpg>

[Paul Oxley]

Larry

I found the GNSS receivers on Amazon.

 https://www.amazon.com/Waterproof-Receiver-Laptop-Interface-Gain/dp/B071XY4R26

Please correct my previous message to show they were from QGP supply (not QCP).

I do not think I have the software. I will look around, but I purchased them some time ago.

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/1910722943.448359.1612905985528%40mail.yahoo.com.

[Paul Oxley]

Wende

That sounds about correct for units that depend on using multiple satellites in different orbits. Hopefully, some if not all the errors can be canceled by using GPS's closely spaced that use the same satellite. 

I looked at the GPS spec. However, it is very hard to translate to this situation.

Paul

https://groups.google.com/d/msgid/sara-list/CC3B1909-6154-4867-8988-38D1C1AEB737%40gmail.com

.

[Marcus D. Leech]

On 02/09/2021 04:44 PM, Paul Oxley wrote:

Wende

That sounds about correct for units that depend on using multiple satellites in different orbits. Hopefully, some if not all the errors can be canceled by using GPS's closely spaced that use the same satellite. 

I looked at the GPS spec. However, it is very hard to translate to this situation.

Paul

https://phys.org/news/2016-02-gps-tracking-centimeter.html

Wende 

To post to this group, send email to sara...@googlegroups.comhttps://phys.org/news/2016-02-gps-tracking-centimeter.html

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/462679380.437773.1612907040578%40mail.yahoo.com.

[Paul Oxley]

Marcus

Sounds interesting. Can you write the software to operate on a small computer? I will give you the two GNSS receivers if you want.

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/602302CF.90105%40gmail.com.

[Captain Anne Flint]

Yes Paul, 

I’ve often found that specs can be vague when a person is trying something innovative with equipment. I suppose you could set up some experiments to determine accuracy. Which would take you sort of tangent to what you were originally doing. So, “in all your copious free time .. “ 

Wende 

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/462679380.437773.1612907040578%40mail.yahoo.com.

[Monroe Pattillo]

If you want to know pointing azimuth minus magnetic influences maybe use a directional GPS antenna oriented to the pointing direction of the scope (GPS ants are typically omni) and use current downloaded satellite ephemeris coupled with trends in active SNR values for rising and setting satellites to determine azimuth.  You’d need to the directional antenna, but it wouldn’t require two GNSS receivers and calculate differential GPS.  And if you have three or more satellites in the field of your directional antenna you could also determine lat/lon.  Altitude above sea level of more accurate using a barometric pressure sensor than determining it through GNSS readings.

 

Monroe

Wende 

An inclinometer sensor did work It depends on gravity for the measurement.

 

Paul

 

 Monday, February 8, 2021, 02:16:53 PM EST, Pablo Lewin <pabl...@gmail.com> wrote:

Thank you Wolfgang 

 

I see that your devices are attached directly to metal and that MAY have a bearing on the Azimuth problem. The strategy I'm using is to attach the sensor to a 1 meter wooden pole /rafter (One could use a PVC pipe) anything  to keep the sensor at least a meter away from the antenna while providing minimum flexure, thus avoiding the potential magnetic interference with the sensor from metal.  I stole the concept from the old MD-80's I used to fly for a living whereby the compass "flux gate" was away from the cockpit and near the right wingtip (Not to be confused with the simple Compass which was located behind the Co-pilot's seat but I digress).

 

Pablo

On Monday, February 8, 2021 at 10:19:03 AM UTC-8 Wolfgang Herrmann wrote:

Pablo,

we have been investigating MEMS devices including the MPU9250 for their usefulnes for our radio telescopes. The finding is that they are sufficient measuring the elevation of smaller telescopes. We actually use them on our 1.2-m and 2.3-m telescopes and on our Ku-Band interferometer for that purpose. What we did find was that the magnetic compass of the MPU9250 was not sufficient for long term accurate positioning in azimuth. The main reasons were long term stability and susceptibiltiy to external magnetic fields.

We started using MEMS some time ago and at that time the ADXL 203 was available. So all our devices in the field are based on this unit For newer installations we will probably consider the MPU9250 or it's successor, the ICM-20948 as the MPU9250 did perform well in the lab (but only for elevation).

--

http://groupsgoogle.com/group/sara-list?hl=en

https://groups.google.com/d/msgid/sara-list/5de3902f-4bcf-4e7f-8f65-0589cd322559n%40googlegroups.com

.

--

sara-list-unsubscribe@googlegroupscom

To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+unsubscribe@googlegroupscom.

To view this discussion on the web visit

https://groups.google.com/d/msgid/sara-list/CC3B1909-6154-4867-8988-38D1C1AEB737%40gmail.com

.

--

--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/462679380.437773.1612907040578%40mail.yahoo.com.

[Paul Oxley]

Wende

At my age, I have the time, but not the physical ability to construct the arrangement to test. That is why I made the offer to Marcus. I would make the same offer to you.

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/E242C0C6-2159-4D15-B7EE-9ABBAA5AEC6B%40gmail.com.

[Paul Oxley]

Monroe.

Thanks for the contribution to the discussion. 

I do not see the reason for directional antennas. The two Omni's would provide the difference in the Lat/Long to use in the differential calculation. The absolute Lat/Long is not of interest. Rather the difference is what is needed.

If we could do it in software, we would prefer to only use one satellite for both GNSS receivers simultaneously to eliminate satellite generated errors.

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/0d8201d6ff2f%24c3e29300%244ba7b900%24%40bellsouth.net.

[Pablo Lewin]

I just performed an experiment with this system. I've attached the video for your perusal. I call your attention to Angle Y axis which depicts 0 as the Zenith (straight up) and angle Z axis as the heading. (Disregard the X axis) 

Now the heading is completely wrong because at the beginning I'm facing South BUT what's important is the heading change seems to be proportionate  which I believe is correct so F I can find a way to calibrate to a known heading then this could be

useful azimuth indication..for $30 dollars..

https://www.youtube.com/watch?v=CrIvUzoczz0&feature=youtu.be

Thoughts?

Pablo (WA6RSV)

[Monroe Pattillo]

Paul,

 

I agree if the goal is a more precise lat/lon positioning.  Differential GPS is I think what surveyors might use these days so we might gain something from the ground they’ve already trodden. 

 

I was referring to mount orientation once position is known and if a directional antenna oriented to the direction the mount were oriented were used it would be able to ascertain the orientation of the mount using only 1 GNSS receiver.

 

Monroe

From: sara...@googlegroups.com [mailto:sara...@googlegroups.com] On Behalf Of Paul Oxley
Sent: Tuesday, February 09, 2021 5:20 PM
To: sara...@googlegroups.com
Subject: Re: [SARA] WitMotion WT901C-TTL mpu9250 High-Precision Gyro+Accelerometer+Angle+Magnetometer,9-Axis Digital Compass

 

Monroe.

 

Thanks for the contribution to the discussion. 

 

I do not see the reason for directional antennas. The two Omni's would provide the difference in the Lat/Long to use in the differential calculation. The absolute Lat/Long is not of interest. Rather the difference is what is needed

 

If we could do it in software, we would prefer to only use one satellite for both GNSS receivers simultaneously to eliminate satellite generated errors.

 

Paul

 

 

 

Wende 

But of course, lol. However, accelerometers measure  the “normal”  or restoring force. And that  is through the centerline of the sensor and with alignment to the earth’s center as you say. But if you take that sensor and move it off angle (ie tilt it, It will change its reading  because the Normal force is now  different due to the angle of it is making.     If solidly mounted to the axis of rotation will have all three axis producing a varying amount of new normal  or restoring forces  For an azimuth drive system to measure the rotation angle, then one of the axes would have to be placed on  something that rotates in accordance with the AZ drive so that a variance if normal force could be determined.  I mentioned that  a bevel gear arrangement with say the x axis accel horizontal and the axle rotated to change the angle of the accelerometer can be used to measure that angular rotation. Of course it would need calibration and signal conditioning, but all doable.  The altitude angular measurement could use a simple one axis accelerometer with output It is, on this kind of mount, just basically a rotator system.  I will look through my  pile of arduino modules and see if there is an accelerometer and if so see if I can gin up a test  rig to demonstrate or show my ignorance, either is possible and likely,  lol.

--

--

To view this discussion on the web visit https://groups.googlecom/d/msgid/sara-list/000001d6feed%24b84719e0%2428d54da0%24%40mayfco.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/462679380.4377731612907040578%40mail.yahoo.com.

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/0d8201d6ff2f%24c3e29300%244ba7b900%24%40bellsouth.net.

--

--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/65138681.476413.1612909204737%40mail.yahoo.com.

[Paul Oxley]

Pablo

In the video, what material is used in the dish? Is it aluminum or steel. Steel would be the worse case test since the magnetic field would be distorted by the antenna.

Paul

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
To view this discussion on the web visit

https://groups.google.com/d/msgid/sara-list/9953d755-11df-45cc-b3f8-f257c1e762e5n%40googlegroups.com

.

[Paul Oxley]

Monroe

Yes surveyors use differential GPS for a first cut. They place one unit on a bench mark and measure the difference to the unknown position of the other. This is likely followed up with the traditional survey measurements using distances and angles for more precision.

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/0db101d6ff35%24721f45d0%24565dd170%24%40bellsouth.net.

[Pablo Lewin]

Hi Paul

Apparently Steel and that's why I mounted the sensor 1 meter away from the dish. Apparently the change in azimuth angle seems proportionate as the dish pans around....

Pablo

[Captain Anne Flint]

Paul, 

Alas, I do not have the time, but thanks for asking. I plan to start building things in about a year. In the meantime, it seems you’ve got some good help. 

Wende 

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/371497124.471937.1612908602630%40mail.yahoo.com.

[Larry Mayfield]

Hmmm, nice thinking. Yet, even the mil system has units have a fairly large finite resolution for location. SO maybe if the units were exactly reading the same when place at the same spot, it might work. I have a GPS module    working right now on my desk in my office. And it does read lat and   longitude to lots of decimal numbers…which change even though my desk isn’t moving, lol. But, at 3 decimal places it is  pretty stable and those are degrees.  But remember a degree at my latitude is pretty wide. So, high accuracy would be needed I am thinking.    I still really think our very best bet is a magnetic rotary  encoder to measure the actual rotations.

 

I did get my 3 axis g measurement and gyro working a few moments ago.   I need to calibrate it now, and then figure out what the dickens the units of measurement are. Doggone code for examples are barely  commented as to what  is going on. Need to do some more research tomorrow.  I’ll let you know what I find about accuracy of GPS. They do altitude but bounces all over  the place with the number of sats above at the time.

 

Buy a couple of gps arduino modules (under 10 bucks ea) and an arduino uno r3 (under 10 bucks also). Get one with an ESP 8266 Web server integrated and you can have it broadcast to you via router as a private network that you can log onto to read the info from you hidey hole . My location is called the Ahem, “The Worm Hole”, but   my wife calls it the worm’s hole…. Fun to mess with and if you buy a starter kit as well, you get a disk full of example codes and how to stuff.  The system you mention might work for the az rotations since the things are flat on the ground But the alt axis is vertical and which moves in direction as wells elevation.  Might be lots of trig involved, spherical types..

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/1910722943.448359.1612905985528%40mail.yahoo.com.

[Larry Mayfield]

They use as many sats as are in view. I have had 12 at one time. Each one gets a Hdop.  I think everything gets averaged together. Not just one.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/462679380.437773.1612907040578%40mail.yahoo.com.

[Larry Mayfield]

Paul, are you older than me? I am 79?

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/371497124.471937.1612908602630%40mail.yahoo.com.

[Paul Oxley]

Larry

I would not suggesting using the concept for elevation angle. This is best done by a incline sensor that uses gravity for the reference.

To use it for azimuth sensing, the two GPS units would need to be mounted on the antenna system in a place that doesn't change with the elevation. One could use a 1 meter stick for the separation.

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/007501d6ff50%2439f1c1e0%24add545a0%24%40mayfco.com.

[Larry Mayfield]

Good morning, Paul!  I am up early nearly every day so I started fiddling around with some numbers. For instance at my latitude right now, which is about 36.168 degrees N then according to an online calculator the diameter of the earth at that latitude: 3959 miles at my elevation. The circumference then is 12438 miles or 65,670 ft, give or take a few.  Dividing that by 360 degrees to get how many feet it takes for one degree = 182418 ft.  a mili degree = 182.42 ft. need micro degrees to get below a foot of resolution.  Micro degree = 0.182 ft.  or  2.182 inches.  So a millionth of a degree is only a tad more than 2 inches that has to be detectable between two different points  (I think, lol.  I am probably all wet here but I think that even a small temperature change can cause things to be off by that  or  more).  amounts).  Let me work backwards.. at one meter radius on the azimuth ring you have a GPS detector.  You want 0.01 degree of accuracy. At 39.37 inches long 0.01 degrees, S=r * Theta and then figure the chord length of the arc.  S = 39.39 inches * 0.01 degrees = 0.3939 inches.  The length of the chord for that movement is Ch = 2*r*sin (1/2*Theta) = 2*39.37*sin (0.5* 0.01 deg) = 0.00687 inches of movement. We have seen, if what I have done is correct (and who knows,  lol) then to detect 0.01 degrees of movement means that a system that can maybe detect 2.182 inches, or a millionth of a degree now has to be able to discriminate 0.000687 inches to get a 0.01 degrees of accuracy.  Hmmm, I don’t  think so…

 

Now I am not so arrogant to think that these numbers are accurate to a lot of decimal places and I certainly invite any and all to point out the errors in my thinking. Especially if you can demonstrate with calculations and experiment using a real GPS unit. I will see if I can do that as well, but, the challenge is out there, lol.  I will see what the GPS unit I have , A NEO-6M module can do for a 1  meter relocation.  Below is the data for 0 displacement, 1 inch displacement and 1 meter displacement of the sensor. All taken within a few moments of each other. Best I can do. Degrees were told to display out to nano degrees (ie 10^-9 decimal positions).

 

0 displacement

 

1 inch (approx.) displacement

 

1 m displacement

 

The GPS/weather data was taken every second. To me it tells the story that it might be ok for finding something  within a circle of say 10 feet or so +-. But quality measurements to that needed to point an antenna to a 0.01 degree or so, is not likely to happen. Micron size measurements are needed to get below a foot of resolution and nano down to sub inches of resolution.  Or at least that is my assessment right now.  I am going to stick with rotary encoder, with better resolution for my viewing pleasure.  I am still trying to figure out my IMU and 3 axis accelerometer values.  That still might offer some hope for me, anyway.  I worked on weapon systems that used IMU devices to hit targets 10000 miles away within 25 feet.  And yet  I am not sure, at this time, whether or not that is even accurate enough, lol.  Back to encoders…

 

No, the data is not in a file somewhere you can down load it.  This was from a snips of a serial monitor from Arduino IDE.

[Jim Abshier]

Paul,

The attached report may be of interest to you since it deals with
attitude determination using baselines on the order of one meter. The
report discusses problems of integer cycle ambiguity resolution and
calibration of antenna phase center migration with satellite look angle.
It does not deal with the multipath effects that can perturb carrier
phase measurements.

Although I don't want to sound discouraging, I think that there are much
easier ways of measuring antenna attitude.

As for surveyors, I suspect that GPS carrier phase differential
measurements are now the preferred approach to surveying.

Jim Abshier

On 2/9/21 5:57 PM, Paul Oxley wrote:
> Monroe
>
> Yes surveyors use differential GPS for a first cut. They place one
> unit on a bench mark and measure the difference to the unknown
> position of the other. This is likely followed up with the traditional
> survey measurements using distances and angles for more precision.
>
> Paul
>
> On Tuesday, February 9, 2021, 05:47:01 PM EST, Monroe Pattillo
> <monroe_...@bellsouth.net> wrote:
>
>
> Paul,
>
> I agree if the goal is a more precise lat/lon positioning. 
> Differential GPS is I think what surveyors might use these days so we
> might gain something from the ground they’ve already trodden.
>
> I was referring to mount orientation once position is known and if a
> directional antenna oriented to the direction the mount were oriented
> were used it would be able to ascertain the orientation of the mount
> using only 1 GNSS receiver.
>
> Monroe
>

> *From:*sara...@googlegroups.com [mailto:sara...@googlegroups.com]
> *On Behalf Of *Paul Oxley
> *Sent:* Tuesday, February 09, 2021 5:20 PM
> *To:* sara...@googlegroups.com
> *Subject:* Re: [SARA] WitMotion WT901C-TTL mpu9250 High-Precision

> Gyro+Accelerometer+Angle+Magnetometer,9-Axis Digital Compass
>
> Monroe.
>
> Thanks for the contribution to the discussion.
>
> I do not see the reason for directional antennas. The two Omni's would
> provide the difference in the Lat/Long to use in the differential
> calculation. The absolute Lat/Long is not of interest. Rather the
> difference is what is needed
>
> If we could do it in software, we would prefer to only use one
> satellite for both GNSS receivers simultaneously to eliminate
> satellite generated errors.
>
> Paul
>
> On Tuesday, February 9, 2021, 05:06:23 PM EST, Monroe Pattillo
> <monroe_...@bellsouth.net> wrote:
>
> If you want to know pointing azimuth minus magnetic influences maybe
> use a directional GPS antenna oriented to the pointing direction of
> the scope (GPS ants are typically omni) and use current downloaded
> satellite ephemeris coupled with trends in active SNR values for
> rising and setting satellites to determine azimuth. You’d need to the
> directional antenna, but it wouldn’t require two GNSS receivers and
> calculate differential GPS.  And if you have three or more satellites
> in the field of your directional antenna you could also determine
> lat/lon.  Altitude above sea level of more accurate using a barometric
> pressure sensor than determining it through GNSS readings.
>
> Monroe
>

> *From:*sara...@googlegroups.com [mailto:sara...@googlegroups.com]
> *On Behalf Of *Paul Oxley
> *Sent:* Tuesday, February 09, 2021 4:44 PM
> *To:* sara...@googlegroups.com
> *Subject:* Re: [SARA] WitMotion WT901C-TTL mpu9250 High-Precision

> Gyro+Accelerometer+Angle+Magnetometer,9-Axis Digital Compass
>
> Wende
>
> That sounds about correct for units that depend on using multiple
> satellites in different orbits. Hopefully, some if not all the errors
> can be canceled by using GPS's closely spaced that use the same
> satellite.
>
> I looked at the GPS spec. However, it is very hard to translate to
> this situation.
>
> Paul
>
> On Tuesday, February 9, 2021, 04:36:34 PM EST, Captain Anne Flint
> <bera...@gmail.com> wrote:
>
> Hi Paul,
>
> I used to work with commercial grade GPS units; if it were me, I’d
> look at the unit/error specs to see if they are accurate to within one
> meter (6 meters just floated up in my brain, but that was a really
> long time ago) and move them out past their error rating[s].
>
> Wende
>
> On Feb 9, 2021, at 3:26 PM, Paul Oxley <oxl...@att.net

> <drm...@mayfco.com <mailto:drm...@mayfco.com>> wrote:
>
> But of course, lol. However, accelerometers measure  the “normal”  or
> restoring force. And that  is through the centerline of the sensor and
> with alignment to the earth’s center as you say. But if you take that
> sensor and move it off angle (ie tilt it, It will change its reading
> because the Normal force is now different due to the angle of it is
> making.     If solidly mounted to the axis of rotation will have all
> three axis producing a varying amount of new normal  or restoring
> forces  For an azimuth drive system to measure the rotation angle,
> then one of the axes would have to be placed on something that rotates
> in accordance with the AZ drive so that a variance if normal force
> could be determined.  I mentioned that  a bevel gear arrangement with
> say the x axis accel horizontal and the axle rotated to change the
> angle of the accelerometer can be used to measure that angular
> rotation. Of course it would need calibration and signal conditioning,
> but all doable.  The altitude angular measurement could use a simple
> one axis accelerometer with output It is, on this kind of mount, just
> basically a rotator system.  I will look through my pile of arduino
> modules and see if there is an accelerometer and if so see if I can
> gin up a test  rig to demonstrate or show my ignorance, either is
> possible and likely,  lol.
>
> Oh, here is a snippet from the web…
>

> *Force Balance Sensor*

>
> <image002.jpg>Force Balance Sensor
>
> These sensors are gravity referenced sensors and are anticipated for
> DC acceleration measurements like ships, vehicles, aircraft and
> seismic events. These sensors are frequently used in inclinometers and
> tilt meters. Force balance sensors are capable of measuring levels
> from 0.0001g to 200g, and the frequency range is from DC to 1000Hz.
> The advantages of these sensors include their high accuracy, a change
> in broad measurement, insensitivity to temperature change and their
> high accuracy.
>
> Larry
>
> Pahrump, NV
>

> *From:*sara...@googlegroups.com <mailto:sara...@googlegroups.com>
> [mailto:sara...@googlegroups.com] *On Behalf Of *Paul Oxley
> *Sent:* Tuesday, February 9, 2021 9:19 AM
> *To:* sara...@googlegroups.com <mailto:sara...@googlegroups.com>
> *Subject:* Re: [SARA] WitMotion WT901C-TTL mpu9250 High-Precision

> Gyro+Accelerometer+Angle+Magnetometer,9-Axis Digital Compass
>
> Larry
>
> The mean of earth's gravity points to the center of the earth. This is
> what is used to sense the elevation angle.
>
> Paul
>
> On Tuesday, February 9, 2021, 09:13:41 AM EST, Larry Mayfield

> <drm...@mayfco.com <mailto:drm...@mayfco.com>> wrote:
>
> Paul, did you come up with a good answer for the sensors? I agree with
> you that mag field sensors are suspect and the earths field changes a
> lot depending on solar winds etc. I still like rotary encoders for my
> project.  Accelerometer might work if a triaxial unit is used.  
> Signal conditioning would be a key element for any of them though.   I
> think any that I use would be a stepper or servo  type motor with an
> encoder on the end of  the shaft to count degrees or pulses and then
> back calculate for degrees  (for data recording) and actual position.
> And of all things, after seeing the example of the sensor being used
> on the  Celestron optical unit, I realized that  I had been thinking

> /inside the box/, lol..  I have been trying to measure the movement of

> the final output when I can measure the input which is easier since it
> urns a lot more than microscopically…
>
> An inclinometer  could even work for an az measuring set up if a gear
> driven unit with a bevel gear   Just need units, with digital or
> analog output for signals, is used.
>
> Back to my second cuppa now…
>
> And thanks for reminding me of the use of inclinometers..
>
> Larry
>
> Pahrump, NV
>

> *From:*sara...@googlegroups.com <mailto:sara-list@googlegroupscom>
> [mailto:sara...@googlegroups.com] *On Behalf Of *Paul Oxley
> *Sent:* Monday, February 8, 2021 3:26 PM
> *To:* Society of Amateur Radio Astronomers <sara...@googlegroups.com
> <mailto:sara...@googlegroups.com>>
> *Subject:* Re: [SARA] WitMotion WT901C-TTL mpu9250 High-Precision

> Gyro+Accelerometer+Angle+Magnetometer,9-Axis Digital Compass
>
> Pablo
>
> I did some experiments with a similar setup. I even tried mounting the
> compass sensor on a non-ferrous (aluminum) angle piece. It didn't help.
>
> Seems to me that the Accelerometer solution would depend on having
> motion present for the measurement. With the antenna motion, it likely
> isn't fast enough to sense this way. The acceleration motion would be
> in bursts when the motion started and ended.
>
> An inclinometer sensor did work It depends on gravity for the measurement.
>
> Paul
>
>  Monday, February 8, 2021, 02:16:53 PM EST, Pablo Lewin

> <mailto:sara...@googlegroups.com>

> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com

> <mailto:sara-list-...@googlegroups.com>

> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the
> Google Groups "Society of Amateur Radio Astronomers" group
> To unsubscribe from this group and stop receiving emails from
> it, send an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

>
>
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sara-list/3190ec43-64f9-460f-af56-526f7c21640fn%40googlegroups.com

> <https://groups.google.com/d/msgid/sara-list/3190ec43-64f9-460f-af56-526f7c21640fn%40googlegroups.com?utm_medium=email&utm_source=footer>.

>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com

> <mailto:sara...@googlegroups.com>

> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com

> <mailto:sara-list-...@googlegroups.com>

> For more options, visit this group at
> http://groupsgoogle.com/group/sara-list?hl=en

> <http://groups.google.com/group/sara-list?hl=en>
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit
>
> https://groups.google.com/d/msgid/sara-list/5de3902f-4bcf-4e7f-8f65-0589cd322559n%40googlegroups.com

> <https://groups.google.com/d/msgid/sara-list/5de3902f-4bcf-4e7f-8f65-0589cd322559n%40googlegroups.com?utm_medium=email&utm_source=footer>

>
> .
>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com

> <mailto:sara...@googlegroups.com>

> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com

> <mailto:sara-list-...@googlegroups.com>

> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sara-list/1243499776.19466.1612826747263%40mail.yahoo.com
> <https://groups.google.com/d/msgid/sara-list/1243499776.19466.1612826747263%40mail.yahoo.com?utm_medium=email&utm_source=footer>.

>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com

> <mailto:sara...@googlegroups.com>

> To unsubscribe from this group, send email to
> sara-list-unsubscribe@googlegroupscom

> <mailto:sara-list-unsubscribe@googlegroupscom>

> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.googlecom/d/msgid/sara-list/000001d6feed%24b84719e0%2428d54da0%24%40mayfco.com
> <https://groups.google.com/d/msgid/sara-list/000001d6feed%24b84719e0%2428d54da0%24%40mayfco.com?utm_medium=email&utm_source=footer>.

>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com

> <mailto:sara...@googlegroups.com>

> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com

> <mailto:sara-list-...@googlegroups.com>

> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sara-list/1040829168.329009.1612891122075%40mail.yahoo.com
> <https://groups.google.com/d/msgid/sara-list/1040829168.329009.1612891122075%40mail.yahoo.com?utm_medium=email&utm_source=footer>.

>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com

> <mailto:sara...@googlegroups.com>

> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com

> <mailto:sara-list-...@googlegroups.com>

> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sara-list/003e01d6ff1c%24f3b08f50%24db11adf0%24%40mayfco.com
> <https://groups.google.com/d/msgid/sara-list/003e01d6ff1c%24f3b08f50%24db11adf0%24%40mayfco.com?utm_medium=email&utm_source=footer>.

>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com

> <mailto:sara...@googlegroups.com>

> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com

> <mailto:sara-list-...@googlegroups.com>

> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sara-list/1910722943.448359.1612905985528%40mail.yahoo.com
> <https://groups.google.com/d/msgid/sara-list/1910722943.448359.1612905985528%40mail.yahoo.com?utm_medium=email&utm_source=footer>.

> <image002.jpg>
>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com
> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+unsubscribe@googlegroupscom

> <mailto:sara-list+unsubscribe@googlegroupscom>.

> To view this discussion on the web visit
>
> https://groups.google.com/d/msgid/sara-list/CC3B1909-6154-4867-8988-38D1C1AEB737%40gmail.com

> <https://groups.google.com/d/msgid/sara-list/CC3B1909-6154-4867-8988-38D1C1AEB737%40gmail.com?utm_medium=email&utm_source=footer>

>
> .
>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com
> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sara-list/462679380.4377731612907040578%40mail.yahoo.com
> <https://groups.google.com/d/msgid/sara-list/462679380.437773.1612907040578%40mailyahoo.com?utm_medium=email&utm_source=footer>.

>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com
> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sara-list/0d8201d6ff2f%24c3e29300%244ba7b900%24%40bellsouth.net
> <https://groups.google.com/d/msgid/sara-list/0d8201d6ff2f%24c3e29300%244ba7b900%24%40bellsouth.net?utm_medium=email&utm_source=footer>.

>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com
> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sara-list/65138681.476413.1612909204737%40mail.yahoo.com
> <https://groups.google.com/d/msgid/sara-list/65138681.476413.1612909204737%40mail.yahoo.com?utm_medium=email&utm_source=footer>.

>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com
> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sara-list/0db101d6ff35%24721f45d0%24565dd170%24%40bellsouth.net
> <https://groups.google.com/d/msgid/sara-list/0db101d6ff35%24721f45d0%24565dd170%24%40bellsouth.net?utm_medium=email&utm_source=footer>.

> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com
> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sara-list/1305466285.493481.1612911473325%40mail.yahoo.com
> <https://groups.google.com/d/msgid/sara-list/1305466285.493481.1612911473325%40mail.yahoo.com?utm_medium=email&utm_source=footer>.

[Larry Mayfield]

Good evening folks,

I have  been fiddling with my starterkitGY521 module which isa 9axis accelerometer, gyro and I think magnetometer. It has all of the processing built into the very small IC.  Here is a you tube link to a fair video on how to access the registers and what they do, but the one that interested me was that the accelerometer axis can be used to find the angles of rotation. Fo rout =r telescoped we need on ly bother with two of the axes  I believe. That is the gyro value and or accelerometer calculation for the yaw around the Z axis and the rotation around o=the x axes for pitch. Yaw equates to the pan  of the base of the telescope and tilt equates to the pitch of the y axes around the x axis.  It can re=measure rotational speeds up to 333.3 rpm  as the high end of the range and if selected thenth elowest range is 46.sometign ropm. )f course, telescopes are very slllooooowwwww.  So the slowest range is the most sensitive for what I want ot do

 

Link

Ep. 57 Arduino Accelerometer & Gyroscope Tutorial MPU-6050 6DOF Module - Bing video

 

Larry

Pahrump, NV

--

--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/000801d6ffd8%244ec35010%24ec49f030%24%40mayfco.com.

[Paul Oxley]

Larry

I got it. You can wire it to a ardruino and rotate  it back and forth quickly.

Now how will it work on a real antenna mount? Will it give repeatable results on all angles? Are the angles accurate when compared with reality? For example, can you point the antenna at the Sun using only its calculated location? 

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/006501d700e1%241867b3d0%2449371b70%24%40mayfco.com.

[Larry Mayfield]

All unknown answers. I am hoping that the unit can resolve at least a 1/10 degree or so although for what I want to do  maybe a half degree is close enough. Trying to figure out the manner of telling the tings where to point and then the arithmetic of the path it has to take with lots of trigonometry and speed changes involved. But they do it for drones and landing after a flight so must  have some accuracy to them. And finding the simple things like resolution seem s almost to be secret, lol.  I will build a small test rig were I can rotate the system at different rates and spans of degrees to see if I can make sense of it.  I am pretty new at this.  I know I can always fall back on rotary encoders.

 

Besides for me it is fun to mess with this  as I never ever did anything like this before.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/796579566.914463.1613098047125%40mail.yahoo.com.

[B & MR Randall]

My thoughts:

·         Using earth magnetic field may not be best for AZ measure.  Whit Reeve could probably give a clue of how much it moves around as he studies geomagnetic field.

·         Grandchildren playing with magnets in area or truck parked too close could alter magnetic north.

·         motors moving dish could be source of unwanted magnetic field.

·         Using 2 GPS's attached to antenna structure could make some amount of RFI at 1420MHz.

·         Accelerometer chips with 2G full scale could work  good for elevation.  A stationary accelerometer reads 1.000 G due to gravity.

·         A mechanical device could translate azimuth rotation to elevation at a 2nd sensor.  Gears would have slop problems.  A cable between 2 pulleys could work.  1950's radio dial cord style.

 

I only do drift scans with azimuth at due north.  Set elevation with 2 wrenches and a level indicator.   I need to be careful to keep track of elevation for data files.  You do have to wait for target object to pass overhead.

 

There are cases where true tracking is needed.  It allows much longer integration times for a pulsar.  Optical photo work needs high accuracy tracking.

 

A shadow of a vertical pole at true local noon will establish a good true north for calibration.    Be very careful with the level and the clock.

 

Bruce Randall

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/007101d70100%24047fef80%240d7fce80%24%40mayfco.com.

[Larry Mayfield]

Good morning, Bruce!

 

I certainly agree with magnetometer issues. I have one that I built for all three axes and have it working. The field does wander around a bit. So I am not counting on mag fields for anything. I love the last item in your post. I have done this in my back yard where I was /am going to put one of my antenna. I will take a photo and send it along later).  This is an old mil manner of finding true north when you are out in the woods of behind enemy lines and your  compass is snafu. I still have my shadow pole stick in the ground and a pair of large (really large) nails driven into the ground for the east west axis. Regards the use of gravity, the manner of using them to track and to provide location direction or degrees for recording for posterity is reasonably easy to do once you have it reading the numbers for g to the highest resolution. The  dinky devices are inertial measurement units and rely on the acceleration of a mass ( I use that term lightly) internal to the IC. The issues are of course are whether or not they can be repeatable and have the necessary resolution. I do not know, lol. But I have one of the GY521 units and will purchase a few more for consistency checks. This is an ongoing experiment in its early stages for me. As I have suggested a pair of 27 dollar rotary encoders with 1024 pulses per revolution may be a far better and quicker manner to do this.  What is going on with me is hobby work to keep my 79 year old brain functioning and having fun in the process.

 

As to stray EMI from motors etc, yeah always a possibility.  Hmm, an interesting thought which I will sooner or later incorporate into my check into this program, is putting one of the GY521 units as working inside my 1420 MHZ feed and seeing if it puts out anything  when inspected with SDR#. If so, that might help with low pass filter design, eh? Or narrow pass band for the real signal?

 

You mention a stationary accelerometer reads 1 g.   Yes, yes it does. But in these little doo hickeys changing  the angle of the axis also changes the g in a predictable manner (trig). The difference between one measurement and the next then is, of course, the rate of change and is the basis of the gyro section or roll pitch and yaw. Since antenna pointing and racking can be slow, it lends itself to perhaps having several of these 2 dollar units located and taking an average to get something usable? I dunno, yet…

 

I have  several large gear boxes and 90 Volt adjustable dc motors that I can use to build a rotary table to try an mimic speed of rotation or at least determine if these sensors are insufficient or not.

 

An ongoing experiment…

 

Thanks for the comments to it all. Great when people chime in with ideas and suggestions and experience! 

 

Larry

Pahrump, NV

 

From: sara...@googlegroups.com [mailto:sara...@googlegroups.com] On Behalf Of B & MR Randall
Sent: Friday, February 12, 2021 4:27 AM
To: sara...@googlegroups.com
Subject: RE: [SARA] WitMotion WT901C-TTL mpu9250 High-Precision Gyro+Accelerometer+Angle+Magnetometer,9-Axis Digital Compass

 

My thoughts:

[Jim Abshier]

Bruce and Others,

I have used sun observations extensively to orient my antennas and lay
out East/West baselines for interferometers. One gadget that I have
found very useful is a home-made sun compass (see attached image). With
the sun compass you can determine North any time of day. Sun azimuth
throughout the day can be obtained using an on-line tool at:

https://www.esrl.noaa.gov/gmd/grad/solcalc/azel.html

The tool reports sun azimuth. Shadow azimuth is obtained by adding or
subtracting 180 degrees. I have written my own program for computing sun
azimuth at 5 minute intervals throughout the day for use in the field.
Solar noon is not the best time for making observations because the sun
azimuth is changing most rapidly at that time. This is especially true
in the summer when the Sun is high in the sky.

Jim Abshier

> *From:*Larry Mayfield [mailto:drm...@mayfco.com]
> *Sent:* Friday, February 12, 2021 12:29 AM
> *To:* sara...@googlegroups.com
> *Subject:* RE: [SARA] WitMotion WT901C-TTL mpu9250 High-Precision

> Gyro+Accelerometer+Angle+Magnetometer,9-Axis Digital Compass
>
> All unknown answers. I am hoping that the unit can resolve at least a
> 1/10 degree or so although for what I want to do maybe a half degree
> is close enough. Trying to figure out the manner of telling the tings
> where to point and then the arithmetic of the path it has to take with
> lots of trigonometry and speed changes involved. But they do it for
> drones and landing after a flight so must  have some accuracy to them.
> And finding the simple things like resolution seem s almost to be
> secret, lol.  I will build a small test rig were I can rotate the
> system at different rates and spans of degrees to see if I can make
> sense of it.  I am pretty new at this.  I know I can always fall back
> on rotary encoders.
>
> Besides for me it is fun to mess with this  as I never ever did
> anything like this before.
>
> Larry
>
> Pahrump, NV
>

> *From:*sara...@googlegroups.com [mailto:sara...@googlegroups.com]
> *On Behalf Of *Paul Oxley

> *Sent:* Thursday, February 11, 2021 6:47 PM
> *To:* sara...@googlegroups.com
> *Subject:* Re: [SARA] WitMotion WT901C-TTL mpu9250 High-Precision

> Gyro+Accelerometer+Angle+Magnetometer,9-Axis Digital Compass
>
> Larry
>
> I got it. You can wire it to a ardruino and rotate  it back and forth
> quickly.
>
> Now how will it work on a real antenna mount? Will it give repeatable
> results on all angles? Are the angles accurate when compared with
> reality? For example, can you point the antenna at the Sun using only
> its calculated location?
>
> Paul
>
> On Thursday, February 11, 2021, 08:48:15 PM EST, Larry Mayfield

> <drm...@mayfco.com <mailto:drm...@mayfco.com>> wrote:
>
> Good evening folks,
>
> I have  been fiddling with my starterkitGY521 module which isa 9axis
> accelerometer, gyro and I think magnetometer. It has all of the
> processing built into the very small IC.  Here is a you tube link to a
> fair video on how to access the registers and what they do, but the
> one that interested me was that the accelerometer axis can be used to
> find the angles of rotation. Fo rout =r telescoped we need on ly
> bother with two of the axes  I believe. That is the gyro value and or
> accelerometer calculation for the yaw around the Z axis and the
> rotation around o=the x axes for pitch. Yaw equates to the pan  of the
> base of the telescope and tilt equates to the pitch of the y axes
> around the x axis.  It can re=measure rotational speeds up to 333.3
> rpm  as the high end of the range and if selected thenth elowest range
> is 46.sometign ropm. )f course, telescopes are very slllooooowwwww. 
> So the slowest range is the most sensitive for what I want ot do
>
> Link
>
> Ep. 57 Arduino Accelerometer & Gyroscope Tutorial MPU-6050 6DOF Module
> - Bing video

> <https://www.bing.com/videos/search?q=arduino+code+for+MPU+6050+9+axis+accelerometer+and+gyroscope&docid=607998410032549104&mid=6C3821B52685F7F734436C3821B52685F7F73443&view=detail&FORM=VIRE>
>
> Larry
>
> Pahrump, NV
>
> *From:*Larry Mayfield [mailto:drm...@mayfco.com]
> *Sent:* Wednesday, February 10, 2021 10:13 AM

> *To:* sara...@googlegroups.com <mailto:sara...@googlegroups.com>

> *Subject:* RE: [SARA] WitMotion WT901C-TTL mpu9250 High-Precision

> degrees (ie 10^-9 decimal positions).
>
> 0 displacement
>
> 1 inch (approx.) displacement
>
> 1 m displacement

>
> The GPS/weather data was taken every second. To me it tells the story
> that it might be ok for finding something  within a circle of say 10
> feet or so +-. But quality measurements to that needed to point an
> antenna to a 0.01 degree or so, is not likely to happen. Micron size
> measurements are needed to get below a foot of resolution and nano
> down to sub inches of resolution.  Or at least that is my assessment
> right now.  I am going to stick with rotary encoder, with better
> resolution for my viewing pleasure.  I am still trying to figure out
> my IMU and 3 axis accelerometer values.  That still might offer some
> hope for me, anyway.  I worked on weapon systems that used IMU devices
> to hit targets 10000 miles away within 25 feet.  And yet  I am not
> sure, at this time, whether or not that is even accurate enough, lol. 
> Back to encoders…
>
> No, the data is not in a file somewhere you can down load it.  This
> was from a snips of a serial monitor from Arduino IDE.
>
> Larry
>
> Pahrump, NV
>

> *From:*sara...@googlegroups.com <mailto:sara...@googlegroups.com>
> [mailto:sara...@googlegroups.com] *On Behalf Of *Paul Oxley

> *Sent:* Tuesday, February 9, 2021 6:11 PM

> *To:* sara...@googlegroups.com <mailto:sara...@googlegroups.com>

> *Subject:* Re: [SARA] WitMotion WT901C-TTL mpu9250 High-Precision

> Gyro+Accelerometer+Angle+Magnetometer,9-Axis Digital Compass
>
> Larry
>
> I would not suggesting using the concept for elevation angle. This is
> best done by a incline sensor that uses gravity for the reference.
>
> To use it for azimuth sensing, the two GPS units would need to be
> mounted on the antenna system in a place that doesn't change with the
> elevation. One could use a 1 meter stick for the separation.
>
> Paul
>
> On Tuesday, February 9, 2021, 08:58:39 PM EST, Larry Mayfield

> *From:*sara...@googlegroups.com
> <mailto:sara...@googlegroups.com>[mailto:sara...@googlegroups.com]
> *On Behalf Of *Paul Oxley

> *Sent:* Tuesday, February 9, 2021 1:26 PM

> *To:* sara...@googlegroups.com <mailto:sara...@googlegroups.com>

> *Subject:* Re: [SARA] WitMotion WT901C-TTL mpu9250 High-Precision

> <mailto:sara...@googlegroups.com>

> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com

> <mailto:sara-list-...@googlegroups.com>

> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sara-list/000801d6ffd8%244ec35010%24ec49f030%24%40mayfco.com

> <https://groups.google.com/d/msgid/sara-list/000801d6ffd8%244ec35010%24ec49f030%24%40mayfco.com?utm_medium=email&utm_source=footer>.

>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com

> <mailto:sara...@googlegroups.com>

> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com

> <mailto:sara-list-...@googlegroups.com>

> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sara-list/006501d700e1%241867b3d0%2449371b70%24%40mayfco.com
> <https://groups.google.com/d/msgid/sara-list/006501d700e1%241867b3d0%2449371b70%24%40mayfco.com?utm_medium=email&utm_source=footer>.

>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com

> <mailto:sara...@googlegroups.com>

> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com

> <mailto:sara-list-...@googlegroups.com>

> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sara-list/796579566.914463.1613098047125%40mail.yahoo.com
> <https://groups.google.com/d/msgid/sara-list/796579566.914463.1613098047125%40mail.yahoo.com?utm_medium=email&utm_source=footer>.

>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com
> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sara-list/007101d70100%24047fef80%240d7fce80%24%40mayfco.com
> <https://groups.google.com/d/msgid/sara-list/007101d70100%24047fef80%240d7fce80%24%40mayfco.com?utm_medium=email&utm_source=footer>.

>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com
> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sara-list/001201d7013a%2463bdc0f0%242b3942d0%24%40net
> <https://groups.google.com/d/msgid/sara-list/001201d7013a%2463bdc0f0%242b3942d0%24%40net?utm_medium=email&utm_source=footer>.

[Whitham Reeve]

Hi Bruce - I agree with you that using Earth's magnetic field is not a good idea. There are too many possible error sources.

Whit

[Pablo Lewin]

After many days of experimentation the elevation portion of this sensor is spot on (Manually, protractor checked at every angle)and stable  so I am now able to accurately set the elevation remotely.

As far as the azimuth situation, the magnetic readout seems to accurately display change in headings and it's pretty stable EVEN though the software does NOT allow a 360 degree readout (Only plus or minus 180 degrees) or any kind of manual calibration.

I have 2 experiments to conduct, 

1) the first one is to set a large compass rose on the base and through a usb camera get the exact accurate readout of the heading (primitive but effective and I have been using this system successfully for one of my HF beams).

2) The second a little more complicated and it will NOT work on cloudy days is to setup a cheap astronomical camera (A mini ZWO120) on the feed itself facing the sky and by coupling it to an astronomical program like  sequence generator pro or sharpcap I'll be able to plate solve the the image thus knowing EXACTLY and within pixels where the antenna is pointing during clear nights, or reasonable close during daytime by plate solving the sun...eventually getting someone to create an ASCOM driver for my USB motor circuits to be able to "follow" a target.

Naturally that won't work on cloudy days and even though the camera will be shielded by the feedhorn itself there's always the possibility of EMI which could easily be fixed by turning off the camera through a usb switch

Even without an ASCOM driver plate solving the sky can give me guidance towards a target..

Naturally this would be  a primitive system but it will keep me busy during the Pandemic and this in itself makes it worth it!

Pablo

[Paul Oxley]

Pablo

Before I would agree, the issue of magnetic interference needs more investigation. 

Perhaps, if you locked the azimuth at a given point, you could probe the measurement by placing large ferrous items (ie. a steel or iron pipe) at various points around the sensors. This would simulate the steel in many of the antennas and mounts. If the measurement changes, not a good solution.

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/a57e9071-0c68-4e9a-a03f-1fe1291c142en%40googlegroups.com.

[Pablo Lewin]

Hi Paul

I understand what you're saying and it makes sense. 

My contention and the thrust of my experiment is not to see whether heading produced by the sensor is accurate, because so far it is not, meaning that it does not correlate with the magnetic heading the antenna is aimed to.

What I'm trying  to find out if the rate of change as the dish rotates at different elevation angles and whether that change is stable and proportionate, by stable and proportionate I mean whether in a steady state position  does it shift? and so far it has not, in other words when it shows 108 degrees it stays there for days when I then  pan to the right and then to the left and then back to the same original position, it still shows 108 degrees.

What I need to determine now is when the dish physically moves 10 degrees that I will see a 10 degree change in the display and for this I need to accurately measure the headings with known positions.

If the above is true then I can build a correction chart that will correlate the erroneous heading displays with the actual headings. After all All I need is an accuracy of + or - 5 degrees for the kind of work I will be performing with this.

I will also use this dish for EME work and for that I can use wide field camera to aim at the moon or even a visual Telrad...

Thank you

Pablo

PS even if I fail ..who cares..It only means that I'm keeping busy and that's good.

[B & MR Randall]

A second compass chip that does not move with the antenna azimuth could be a reference to correct as the magnetic field of the earth changes.  A lookup table in software could correct for magnetic material in the dish.  This might be reasonably accurate and be a simple system from a mechanical point of view.  Magnetic sensor on a small circuit board is less than $10 from Adafruit or Sparkfun.

 

Added benefit of reference compass is you could record results and observe magnetic fields.

 

Bruce Randall

 

1 inch (approx.) displacement

 

1 m displacement

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/cb61526a-7582-49da-b474-508042600a97n%40googlegroups.com.

[Pablo Lewin]

Thanks Bruce, that's an intriguing possibility. I'll look into those magnetic sensors next..

Pablo Lewin

[Paul Oxley]

Pablo & Bruce

When I attempted to use a magnetic sensor on a steel dish, it proved to be impossible to use the data. The steel in the dish would distort the sensors with changes in either the azimuth of the dish or its elevation.

The elevation of the dish can be measured by any sensor which uses the earth's gravity. This worked with a simple Inclinometer chip in my setup. It could be done with other solutions.

The azimuth of the dish could be measured with various schemes of monitoring the amount of rotation. This method requires careful calibration of the rotation measurement. It is also limited by the granularity of the rotation measurement. The easy solutions are usually limited by granularity. We need to be able to sense an azimuth to <1 degree for most RA applications. The use of the beamwidth for this limit is misleading since we need to be near the center of the beam to avoid a undesirable loss of sensitivity at the 3 dB down points on the beamwidth.

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/64e2081e-83f3-4954-8e93-e363d00fa55bn%40googlegroups.com.

[Hamish Barker]

forgive my ignorance, but why the fascination for using inherently imprecise and subject to interference magnetic sensors rather than simple optical encoders? A $15 encoder with gearing down can provide 2400 pulses/rotation, so about 1/3 degree precision (assuming +/- 1 pulse error.).

Is this for a fixed or mobile installation?

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/1086389309.1523611.1613274441083%40mail.yahoo.com.

[Paul Oxley]

Hamish

I like your solution. It is both simple and has the necessary precision. 

Could you provide more details on the method used including the rotation reduction and sensors?

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAAdNG_JMivfdr-15T9-m-QO47Zku5yE5K60F7y2mt369O7jXYQ%40mail.gmail.com.

[Pablo Lewin]

What software exists (windows) for such an encoder? I haven't been able to find a software solution for the inexpensive encoders. Could you give me vectors to type of encodr/software combination?

thank you

Pablo

[Hamish Barker]

The link in my previous message referring to the cloudy nights astronomy forum contains details of the hardware, processor and software. Total cost at the time was about usd40. 

The system provides a WiFi web server which responds to a simple html query with encoder count. But it could also be programmed to simply spit out the current encoder count every 5 seconds or whatever you want, into serial port. Happy to help alter the arduino code (actually it runs in an esp32 something, but it's programmed via the arduino suite). 

Gearing uses 2mm pitch tooth timing belts which are cheap and easily available at maker suppliers or ( my preference) aliexpress. Shafts and bearings and pillow blocks similarly. They are used in 3d printers hence the easy availability.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/457901992.1532860.1613275105550%40mail.yahoo.com.

[Larry Mayfield]

Pablo, I love what you are doing…

 

Would you share the code work you are doing to make things happen? I ask because right now I am using a cheap (and I mean cheap, lol, just purchased 10 of them for under 10 USD) Module, a GY-521, to figure out how to do this. The AZ part seems to be reasonably easy to do as it is just a unit circle and the G’s it produces on the x and y axis as it rotates can be tailored (I think) to match any drive needs for that axis. If the Y axis is the alt axis then that should be fairly easy to do as well since one leg of the pointing angle (ie Z axis, is constant at 1 g) then as the Y axis rotates (and the gyro angle is known the dish pointing alt angle can be determined.   Right now my numbers are bouncing all over the place.  But mostly there are in the 100’s of a degree range (I think, lol). It is difficult to get to the units used in the darn things and those who know are reluctant to share what they know, I guess.  My project is a long one as being 79 years old and non-electronics oriented is hampering me, but that is part of the fun…learning new stuff…  My thoughts are that I want to mount the sensor centerline of the x, y and z axis at the very center of rotation of the dish and it would be mounted perfectly flat and parallel the axes.  Then the transformation math is pretty easy (hopefully, lol) and is just the arctangents of the G levels at the measurement time for the axes.  The acceleration of the axes is straight forward as one considers that they are flat to the earth and then only the x and y axes are needed to find those values to get the angles. The y axis with its motion and the rate gives you the position that the measurement is made and with the g level of the y and z  axes the pointing angel can be found. Or so it says in my very old brain‘s consideration…  I have a pile of small geared dc motors and some adjustable speed controllers so I will try and build something to demonstrate how my thinking can be accomplished.   Keep up the experiments on the how of it all!

 

Oh, and I think that with an alt az (tilt pan) drive system, all you need is 180 degrees in each axis to cover the entire sky.

 

I think my email addy is on my profile on the SARA list, interested in the how of what you are doing!

 

Many thanks,

 

Larry

Pahrump, NV

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/a57e9071-0c68-4e9a-a03f-1fe1291c142en%40googlegroups.com.

[Larry Mayfield]

I agree, lol..  And that is my fallback position, lol. However, trying to do this is part of the fun and learning and perhaps making it even cheaper. For instance the  6 dof freedom, but with magnetic input capability can be purchased for under 2 bucks in a 10 lot quantity.  So this is partially trying to get it all down to one module based system with just code work to make it go.  I think. At least that is what I am trying to do with the cheap sensors I have. It can be done without mag sensors if aligned to earth rotational axis for the Y value.   I think that encoders are the quickest implementation strategy, yet it is satisfying to try and figure out different and perhaps even better ways. There are new encoders available, magnetic encoders that work in the same general way as the optical ones or capacitive ones, but more accurate (or so I am told). I suspect that thy us similar techniques to what is being tried now.

 

It would be an easy solution to do both on the dame dish for comparison I think. Pablo? Can you pair up rotary encoders as a comparison to you set up and see what the delta values are?  Seems an obvious thing to do… I will consider that also for any test rig I fab up.

 

Larry

Pahrump, NV

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAAdNG_JMivfdr-15T9-m-QO47Zku5yE5K60F7y2mt369O7jXYQ%40mail.gmail.com.

[Larry Mayfield]

Hamish, could you repeat the link  to the cloudy nights location? I musta been asleep in classs that day…

 

Larry

Pahrump, NV

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAAdNG_JYhyK5pq1nVxOzpUo_uDFdSN4QwiknxemgKhmdsR0zKw%40mail.gmail.com.

[Hamish Barker]

https://www.cloudynights.com/topic/589521-37-dobsonian-dsc-for-diy-makers/

At first magnetic sensor was used but later went to 1 encoder then 2

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/001c01d702e5%2450f0e590%24f2d2b0b0%24%40mayfco.com.

[Pablo Lewin]

Hamish

I believe the system that you propose is a WIFI system (Picture above from the article itself). I purposely avoid using WIFI near my dish and in fact the HEADLESS control computer running my system is mounted ON the dish mount itself and has a 3.5ghz  clock to be as far away from 1.4ghz as possible and NO WIFI Bluetooth  but an ethernet cable.

Can you show us the system you implemented on your Dish so that I can get an idea?

Thank you

Pablo Lewin

[Pablo Lewin]

I found the following motor which would easily be able to move my dish (I've been using a cheap geared 15 dollar amazon for over a month with zero problems) This motor comes with a built in encoder. I will ask the Pololu engineers for a possible software solution

" This gearmotor consists of a low-power, 12 V brushed DC motor combined with a 377.93:1 metal spur gearbox, and it has an integrated 48 CPR quadrature encoder on the motor shaft, which provides 18,140.79 counts per revolution of the gearbox’s output shaft. The gearmotor is cylindrical, with a diameter just under 25 mm, and the D-shaped output shaft is 4 mm in diameter and extends 12.5 mm from the face plate of the gearbox. "

https://www.pololu.com/product/4890

[Hamish Barker]

ah yes of course, silly me, wifi would not be helpful here! I used the system on my telescope. My H1-observing 90cm dish relies on sneakerdrive(TM). (i.e. go out in the yard and move it to a new position.). 

 You could use an arduino without wifi, and just use the USB to the arduino to interrogate and respond as serial text (or forget the interrogation and just have the arduino spit out current alt and az encoder positions every 5 seconds to its serial port. (i.e. " alt = 5329, az = 1248 /r" and your system would then need to parse that simple string. Add error codes etc as you like. That would make the code even simpler than the one done to date. 

If you want the arduino to instead provide an output of current degrees az and alt, then some sort of home/limit switches would also be required and do a startup turn until finding home switch, then be able to do the maths from there. there are plenty of I/O pins on an arduino to support a couple of extra switches.

I think this would be much more reliable than using magnetic headings.

Note that the arduino can actually be quite some distance from the encoder but you would need a cable with sufficient conductors (and appropriate multipin connectors), to the encoders. this might be more trouble than rigorously shielding the arduino in a metal box with well shielded serial and power cabling in/out.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/576dcd90-92fd-4758-960b-dd58f8b0f521n%40googlegroups.com.

[Larry Mayfield]

Nice set up! ESP 8266 as a web server.  It is what I have in mind for my rig when I get to my final homestead location, if ever.   Now they can be had fully   integrated with  the UNO R3 and the Mega2560 MC units. Along with a Ch340 serial UART system. Switch selectable. Each piece can be run separately or all lot once.

 

Now I have  one of my dumb questions; number 4893 to be  exact: when noise or  microwave is intercepted by the backside of a parabola dish, is that signal reflected elsewhere and out of play for the dish? And if the dish is  grounded well does that help reduce noise? Seems to me that when I am powered up one day in the future, that is a tiny experiment I want to run. Power up my test system, in the open air behind the dish and take a look at the noise that comes though and is displayed on SDR#.  What should I expect to see?  Then I would repeat on the front side of the dish and that way I could see the frequencies that I need to filter out of the signal being received specifically. And of course, would repeat all of the above with electronics encased in metal enclosures.  Then what about a screen fence attached to the front face circumference of the dish, perhaps as tall as the focal point, to reduce any side signals coming into the dish  Fence might be angled outward sufficiently to prevent any reflections from it from getting to the parabolic surface.

 

Wow, this is more than one question..    feel free to ignore, lol…

 

Larry

Pahrump, NV

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/83f3d3aa-44d7-4feb-a752-a3fc5e5a0c8bn%40googlegroups.com.

[Larry Mayfield]

Hamish, many thanks for resending this. So this set up is for both axes of a tilt pan (Alt-Az) type system.  I like this. Much better than what I have been trying to accomplish.  I will incorporate this into my planned antenna farm. I have a 3 meter dish, a 2 meter dish, 2 Dish TV dishes, I purchased way back when, and a dish that was used to send my internet microwave signal to my house from a tower (they went under as a business so I inherited it).  

 

Does the code include that needed for the GY 511 unit and the rotary  encoder? I think you said it did.  This is one of the issue I am currently trying to sort out on my GY 521 unit. I want to be able to record the az, alt, the time of day,  date,  lat, lng, ambient temp, humidity, baro pressure, altitude,  and ultimately, the radio signal as processed   

 

Oh, I see that you eliminated the magnetic encoder and replaced it with a rotary as well?  So two rotary encoders? 

 

And since I started this email yesterday, I have now taken a look at eh GitHub data you posted. The *.ino file has a wealth of information and is well commented. I am impressed with the work shown in that code, lol. WOW, pile of work there.  I will be working though it line by line. From the comment in yellow below does encoder 2 work like number 1?  Just a duplicate except for the axle orientation?  I may try this later with an ESP32 web server because it has 32MB of flash memory, for maybe web server html code.

 

Again, many, many thanks for sharing your information!

 

You, Sir, are my hero!

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAAdNG_%2Ba8nzO7iLC2Su%3Dys5w67%3D3v7cxe4U0kH8-W_jybiRfgw%40mail.gmail.com.

[Pablo Lewin]

Well after having experimented full time with the  WitMotion WT901C-TTL mpu9250 High-Precision Gyro+Accelerometer+Angle+Magnetometer,9-Axis Digital Compass sensor for 20 days straight and after having learned to do a FULL calibration with it installed in place (Two types of calibrations full which needs to be removed from the mount and ONLY done once and then the partial which takes literally seconds with the sensor installed in place) I can safely say that IT WORKS!.

The software it comes with is easy to use, the Elevation display is pretty accurate, the azimuth display goes from +179 degrees to -179 degrees so it takes a little computation to get the desired heading so for example an azimuth of 210 magnetic would display as -30 degrees on the display because after  +180  it starts at minus numbers so -30 equals to 210 degrees so on so forth.

Keeping in mind that my 2.1 meter dish beam width at 1.4.Ghz is 7 degrees..this is good enough for this system. 

Now every time I point the dish at bright portions of the Galaxy no matter the time of day, I get rewarded with a strong signal.

Is this little $34 dollars sensor research quality? absolutely NOT!, but for a dabbler/noob like myself, it allows me to get some fun and interesting results when used in combination with Radio Eyes..

It's not perfect but it works...mileage of course may vary...and it will...

Pablo Lewin (WA6RSV)

The Maury Lewin Astronomical Observatory

Glendora CA. USA

[Larry Mayfield]

Nice work Pablo!

I am currently doing something similar, but way behind you. I amusing an MPU 6050-DMP IC as installed to make a GY-521 6DOF with digital motion module. It is like pulling teeth to find the correct manner to get the correct offsets for the axes and then to calibrate then for the as installed position.  But I am chipping away at it. Being older than dirt isn’t helping me though.  

 

Glad you have it all working now!!

 

Larry

Pahrump, NV

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/dfd81a62-ad30-466d-af00-ff176af70d01n%40googlegroups.com.