Vacuum Sensor for Head

[Friedrich Mäckle]

I am looking for a vacuum sensor for pickup sensing.

So my first guess would be:

• Digital Output (Ideally I2C with 2 addresses available)
  
• Smoothie compatible (Would be great to just use I2C and write a software module for the smoothieware)
• Differential
  
• Differential Pressure Range < 75kPa (Max pressure will be the pump to environment, but I can only guess what is measured at the pickup head, i guess maybe something around 30kPa?)
  
• Preferrably SMD
  
• Low Cost (<30€)

I think its best to position the sensor very close to the nozzle, so on the head itself, I would like to have two sensors there, but of course voltage drop and other problems come in there with ~2m of wiring).

Any thougts or measured pressure values on this? What models are you using?

Best, Friedrich

[Arthur Wolf]

On Sat, Aug 5, 2017 at 2:45 PM, 'Friedrich Mäckle' via OpenPnP <ope...@googlegroups.com> wrote:

I am looking for a vacuum sensor for pickup sensing.

So my first guess would be:

• Digital Output (Ideally I2C with 2 addresses available)
  
• Smoothie compatible (Would be great to just use I2C and write a software module for the smoothieware)

communication over i2c won't work for the lengths between the smoothieboard and the head.

you want at least rs485 there.

• Differential
  
• Differential Pressure Range < 75kPa (Max pressure will be the pump to environment, but I can only guess what is measured at the pickup head, i guess maybe something around 30kPa?)
  
• Preferrably SMD
  
• Low Cost (<30€)

I think its best to position the sensor very close to the nozzle, so on the head itself, I would like to have two sensors there, but of course voltage drop and other problems come in there with ~2m of wiring).

Any thougts or measured pressure values on this? What models are you using?

Best, Friedrich

--
You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+unsubscribe@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/080af725-cbbd-443b-88af-e60fbe0fec7f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Courage et bonne humeur.

[Friedrich Mäckle]

Yeah, you are most likely right, but I could still use a controller and send the values to either the Smoothieboard or the PC

Okay direct attached could be tricky, may I ask what setup concerning this you are running?

[Friedrich Mäckle]

Well to simplify this, could everyone who has a Vacuum Sensor just post the Part Name here?

I could summarize that later into some kind of knowledge base later.

[Michael G.]

I'm using NXP MPXV6115VC6U sourced from digikey.

[Trampas Stern]

I also use the MPXV6115VC6U, it is a 5V part where my processor is 3.3V so I added a voltage divider on the output.  However you can directly interface it to an Arduino Uno. 

My plan was to use one for each nozzle and one for the vacuum source, to detect errors with solenoids. However I might change the third one to use a pressure sensor, that is use a second vacuum pump (or same one) as pressure source but then PWM modulate the power to pump to generate the correct pressure for puff of air to release part from nozzle, I might need to add a coffee can pressure tank (capacitor for air) to help with regulation. 

Trampas

[TheCunningFellow]

MPXV5100DP

DP is dual port so can be vacuum and pressure sensor depending which way its plumbed in.

[Mike Menci]

Hello Trampas, 

I'm in process of connecting this   MPXV6115V sensor to Smoothie - can you please update me on functionality of the sensor and values of resistors in voltage divider ?

Thanks Mike

[Trampas Stern]

I did not hook up to smoothie but for resistor divider if you use a resistors with such that R1/(R1+R2) = 8/12 you will convert the 5V to 3.3V.  For example R1=8k and R2=4k, this will put the 5V sensor output to 3.3V.  Note that the internal resistance of the sensor might come into play, so you might need larger resistors. 

I also never finished my build with the MPX6115V sensor as I ended up getting a Manncorp PnP. 

Trampas

[ma...@makr.zone]

I did not hook up to smoothie but for resistor divider if you use a resistors with such that R1/(R1+R2) = 8/12 you will convert the 5V to 3.3V.  For example R1=8k and R2=4k, this will put the 5V sensor output to 3.3V.  Note that the internal resistance of the sensor might come into play, so you might need larger resistors.

Because of the Smoothieboard input pullup resistor of only 4.7k plus 10µF capacitor, a simple resistor divider will not do, or it would have to be so low value that the MPXV6115V absolute max output current of only +/-0.5mA is violated.

Look here for a solution. 

https://makr.zone/vacuum-sensor/192/

My chip was a slightly different one but the MPXV6115V would work the same, just with a wider vacuum range.

_Mark

[Marek T.]

You can also use just simple npn transistor in Open-collector mode. Emitter to the ground, base throught resistor to the sensor output and collector to the input of the Smoothie. And little play to tune the value of the resistor to set transistor in reguired part of his characteristic, really not philosophy.

[ma...@makr.zone]

You can also use just simple npn transistor in Open-collector mode. Emitter to the ground, base throught resistor to the sensor output and collector to the input of the Smoothie. And little play to tune the value of the resistor to set transistor in reguired part of his characteristic, really not philosophy.

Did you mean PNP instead of NPN?

See PNP emitter follower.

Yes this works, but it is not that much simpler. You are about 5-10 times slower on rising pressure/broken vacuum (to detect a mis-pick). You also get a temperature dependant voltage offset:

Simulated with different resistors:

You still need the +5V for the sensor. You still need local bypassing. You still need the voltage divider. With all these parts you still need a PCB of some kind. 

So...

_Mark

[Marek T.]

NPN not PNP. No need extra divider then.

Emitter to the ground of Smoothie.

Collector to the thermistor input of Smoothie. Collector will pull down from 3.3V (through internal Smoothieboard pullup resistor 4k7) to 0V (A-GND or if you like GND of Smoothie). "Automatical divider" :-)

Base through resistor to the sensor output.

I have this done this way.

Only instead of typical NPN I've used NPN transoptor to opto-isolate A-GND of Smoothie from GND of Sensor.

Maybe it has not some ideal characteristic perfectly following sensor output but works good enough and fast enough.

[Marek T.]

Ok, your PNP good as well of course. I just used some opto (PC817 probably) with emiter output that's why suggested npn...

Hey, I don't tell my circuit is better, your one is very good and I wanted do similiar untill found that must separate the grounds. Then used opto and when it worked I stopped further thinking :-).

[ma...@makr.zone]

Hi Marek

with PNP it will be an inverter:

The slowness is reversed: fast vacuum break detection slow vacuum build detection.

Without the divider you will lose some of the sensor range, see the ramp starting at 0.8s. Which is probably OK as you never reach that negative pressure anyway.

_Mark

[ma...@makr.zone]

Sorry meant "with NPN" of course.

[ma...@makr.zone]

Need to step to higher R values to get some "linearity", but this will be quite temperature dependent, won't it?

[Marek T.]

Probably it will be dependent but the temperature is not changing over 10 deg I think.

I haven't tracked how it is exactly with the transistor as I told.

In my machine there is a pressure-to-vacuum converter (micro ejectors) with some rising/break time not very very fast (~60-80ms). When I have been comparing edges on the output of the sensor and output of the transoptor they were very close each other.

[Mike Menci]

Hello gents., 

Thank you for your time - I am on trip and will study this a bit later!!

Thanks

Mike

[Marek T.]

Pleasure. Mark's signals lab simulations are impressing for me :-). And for sure his project is better except of missing grounds separations (what was necessary for me but maybe not important for you).

[Trampas Stern]

Or just a resistor divider and a voltage follower op-amp.... 

[ma...@makr.zone]

Or just a resistor divider and a voltage follower op-amp...

Yes, of course.

Just to explain: Like I said in my post, I just had a dual opamp laying around. There is no smaller DIL package than 8-pin, so dual opamps are the natural choice to stock, as they cost (almost) the same. An unused opamp in a package must be properly connected too (see image below). On a strip board where it's easiest IMHO to bridge traces using SMD parts, this seemed more complicated than buffering twice. Also this way you get a more "universal" circuit as it actually outputs a strong signal at both the +5V and +3.3V levels.

https://makr.zone/vacuum-sensor/192/

_Mark

[Trampas Stern]

Another option is to just get a Arduino Nano/uno/etc.  (<$5) and connect it up to the ADC.   That is the Nano/Uno are 5V parts and you can just set up a Mxx command through USB serial and just have OpenPnP read the sensor using the Arduino. 

https://www.amazon.com/Arduino-Emakefun-board-ATmega328P-Micro-controller/dp/B0775XQXRB/ref=sr_1_1_sspa?s=pc&ie=UTF8&qid=1528893817&sr=1-1-spons&keywords=Arduino+nano&psc=1

The firmware from my board: http://misfittech.net/16-channel-fet-pwm-driver/ can be used as a starting point... 

Using the Arduino can provide the 5V for sensors and directly read them. 

Trampas

[ma...@makr.zone]

If you really want it super cheap then just don't use the thermistor inputs, but use one of the spare Smoothieboard pins on expansion headers that have ADC capabilities. Add voltage divider and cap close to pins.

Well, I assume Smoothieware does allow you to configure any pin as a temperaturecontrol pin, though I haven't tried.

Spare ADC pins:

AD4 P1.30   
AD5 P1.31   

Normally used for click button and buzzer on a GLCD display.

http://smoothieware.org/lpc1769-pin-usage#lpc176x-adc-channels-and-pins

_Mark

[Marek T.]

What makes that it's cheaper solution than using thermistor inputs??

[ma...@makr.zone]

Are you the same Marek T. that proposed this to drive the large input cap? That's why.

Well I think I proposed a workable solution that is robust and still reasonably simple to make with common parts and is compatible with Smoothieboard's or any other similar controller's inputs. I'm not saying its the super-cheapest or dirt-simplest or "hackiest" solution ever. That was never the goal.

_Mark

[Marek T.]

It's me but I've told do decrease original capacitors not drive large.
But can't see money saving when using other port than thermistors port.
Anyway, no matter, what about money we talk here, is it difference if it costs all together $5 or 15?