Utracer 6 problem

[Sándor Filotás]

Hello!

I have a Utracer 6 device, unfortunately I am not satisfied with it when a very accurate device works. But it's the second time T65 Fet has failed, the transistors on the anode and screen sides are constantly failing T63 and T83, has anyone experienced this? I'm very disappointed, the device is also unstable, I start a measurement, Hv led on starts and freezes.... It has to be restarted many times, now it's not good, both FETs on the anode side are broken, and so are the transistors.

[Ihor Smal]

Do you have a standard build without any modifications, so everything is according to the manual, and do you use a cable connection to the com port? I used to have similar behavior while using cheap Bluetooth dongles for com port many years ago. 

On May 23, 2023 18:29:58 Sándor Filotás <sturmg...@gmail.com> wrote:

Hello!

I have a Utracer 6 device, unfortunately I am not satisfied with it when a very accurate device works. But it's the second time T65 Fet has failed, the transistors on the anode and screen sides are constantly failing T63 and T83, has anyone experienced this? I'm very disappointed, the device is also unstable, I start a measurement, Hv led on starts and freezes.... It has to be restarted many times, now it's not good, both FETs on the anode side are broken, and so are the transistors.

--
You received this message because you are subscribed to the Google Groups "uTracer" group.
To unsubscribe from this group and stop receiving emails from it, send an email to utracer+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/utracer/6abff9e8-96cf-4dfc-93cb-a2cf90b6d095n%40googlegroups.com.

[Sándor Filotás]

Hello!
 
Usb ttl converter, cp2102, but sometimes it works fine for a week, then I can't start it... I don't understand, the other bigger problem is the broken parts, I haven't changed anything on the switch, everything is factory default. Of course, it is almost impossible to get the fet, I will be curious to see how we can solve this

[Ihor Smal]

And that TTL converter is connected directly to the PIC controller Rx/Tx pins or still to the com port's max232 chip?

My gets were also burning but it is just because the communication gets lost and high voltages are still present and running through the tube. It does happen because of the unstable communication with the PIC controller, believe it or not and not because of the circuit design around the fets. Try standard serial cable, I have never had problems with it. Also when the HV led hangs, you should not just switch uTracer off and on, you do have to issue a special command and that can be done just by staring and stopping the measurement cycle, or with uTracerJS software just sending 30.... Command from debug tab. 

If your cable sometimes works and sometimes not it is already indication about the poor communication due to the cable, the serial port on PIC is stable by design and always works. 

On May 23, 2023 19:24:54 Sándor Filotás <sturmg...@gmail.com> wrote:

Hello!
 
Usb ttl converter, cp2102, but sometimes it works fine for a week, then I can't start it... I don't understand, the other bigger problem is the broken parts, I haven't changed anything on the switch, everything is factory default. Of course, it is almost impossible to get the fet, I will be curious to see how we can solve this

Ihor a következőt írta (2023. május 23., kedd, 18:49:51 UTC+2):

Do you have a standard build without any modifications, so everything is according to the manual, and do you use a cable connection to the com port? I used to have similar behavior while using cheap Bluetooth dongles for com port many years ago. 

On May 23, 2023 18:29:58 Sándor Filotás <sturmg...@gmail.com> wrote:

Hello!

I have a Utracer 6 device, unfortunately I am not satisfied with it when a very accurate device works. But it's the second time T65 Fet has failed, the transistors on the anode and screen sides are constantly failing T63 and T83, has anyone experienced this? I'm very disappointed, the device is also unstable, I start a measurement, Hv led on starts and freezes.... It has to be restarted many times, now it's not good, both FETs on the anode side are broken, and so are the transistors.

--
You received this message because you are subscribed to the Google Groups "uTracer" group.
To unsubscribe from this group and stop receiving emails from it, send an email to utracer+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/utracer/6abff9e8-96cf-4dfc-93cb-a2cf90b6d095n%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "uTracer" group.
To unsubscribe from this group and stop receiving emails from it, send an email to utracer+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/utracer/e7580d0c-dd8e-4af0-ab21-74e947651f54n%40googlegroups.com.

[Sándor Filotás]

Hi, it is directly connected, I took 232 out of the socket.

To view this discussion on the web visit https://groups.google.com/d/msgid/utracer/18849ac4e28.27fb.16a52cc106b662770bf54050d014535e%40gmail.com.

[Benjamin Jordan]

I'm having similar issues. 

I just finished the kit, and things tested fine most way through the manual test procedures.

But when I try to run the test I(Va=Vs, Vg) with Vh fixed - communication is lost. 

It seems to me the PCB design is such that there's a LOT of EMR and that interferes with communication. I was using a similar USB-TTL Com port device and had it connected directly to the PIC. 

But I still have other issues... I'll post questions in another thread about it.

I've had one of the main SiC FETs de-solder itself from heat because the PIC froze and left it switched on. Luckily, the internal RDS(on) of these is high enough that it just got really hot but didn't get destroyed. I have replacements.

This product needs a serious PCB re-design by someone who knows how to design them for switching PSUs. I hope Ronald puts a firmware update with a watch-dog timer in use so that if the PIC loses communications it shuts down every boost converter to prevent destruction...

[Sándor Filotás]

Hello!

Unfortunately, I ended up with a short circuit on the PIC power supply. I consulted with them, it ended up being mailed back to the Netherlands, supposedly I soldered the T80 in carelessly, that's why the boost converter didn't always start, Ronald put a resistor in another place.
Since then, I have used it a few times, it only froze once, but its software could also be improved, not only in terms of hardware.

[Sándor Filotás]

The translation is wrong, I didn't cause the short circuit, it just went wrong...

bj.a...@gmail.com a következőt írta (2023. július 28., péntek, 6:17:10 UTC+2):

[Benjamin Jordan]

I had similar issues. 

I had been careless somehow and sent the kit back to NL for Ronald to fix my mistakes. He's a true gentleman.

It's on the way back to me now. 

Turns out in my enthusiasm to get it working I was rushing. I did somehow fry the current sense opamp (I must have inserted it wrong in the socket I can not think of any other explanation).

Ooops! I should know better. 

[Never Mind]

In my experience this never worked. PIC blowing up and strange operation was the result.

What worked perfectly and has continued to do so for years now is put the  max232 IC back in the socket.

Purchase a true USB to TRUE Rs-232 with a real a DB9 connector using a real FTDI chip. No other chip. Spent the extra money for the real thing.

I agree none of this should matter, the cp2102 connected directly should be fine BUT IT IS JUST NOT how it is.

There is something wrong with the PIC serial port and how this PIC issue interacts the cp2102 and Ronald's firmware that causes the firmware to go crazy and damage the board.

So to stop the suffering just get a real USB to real FTDI chip with a real DB9 connector and put the max232 IC back in the socket. All was then good in my experience.

[Davo]

Lol isn’t this the first thing you have to with the uTracer…

Ronald mentioned this many times though!

I had 1 USB port for around 20euro’s inc. Shipping!

Op 23 aug 2024 om 21:46 heeft Never Mind <bobblue...@gmail.com> het volgende geschreven:



To view this discussion on the web visit https://groups.google.com/d/msgid/utracer/092a1d3f-3894-4b1e-b420-d5fab29770b8n%40googlegroups.com.

[D.A.R Achterberg]

Ps. the logilink has the FDTI chipset

I found that the prices with an FDTI quite varies, much higher then 20euro’s though i always make money on prices, I do that on Ali also

Op 23 aug 2024 om 23:35 heeft Davo <djda...@gmail.com> het volgende geschreven:

Lol isn’t this the first thing you have to with the uTracer…

Ronald mentioned this many times though!

I had 1 USB port for around 20euro’s inc. Shipping!

<image0.jpeg>

To view this discussion on the web visit https://groups.google.com/d/msgid/utracer/94CD538E-0610-4FD8-9369-0F5A3301BA78%40gmail.com.

[Never Mind]

Ronald in my opinion should update his documents to remove any and all suggestions on using a direct TTL level RS-232 interface to a utracer.

I just causes problems and suffering in too many uTracer users. Why he has left this in the documents to mislead his users for so long is a puzzle to me.

The direct to PIC serial port TTL RS-232 connection is at best fragile and often will cause mis-opperation or damage to the uTracer hardware.

If you use the direct TTL RS-232 connection you are connecting to a naked PIC serial port. This is simply poor engineering practice to connect long external cables to any naked micro controller port.

There will be ZERO protection for the PIC's serial port from static discharge, RF interference on the cables, ground bounce between the PC and uTracer, any noise pickup or over voltage when the uTracer is powered down and the cable is left connected.

Yes it will work some of the time in laboratory development situations however it is just not a recommended way to connect long cables to any micro controller based product.

The direct connection is especially scary with a uTracer that generates both high voltage and high current pulses. Think little local lighting storm.

When using a proper RS-232 interface chip like the max-232 and a REAL USB to DB-9 connector interface cable none of the above issue will be present.

[Martin Manning]

I have been using an FTDI TTL-USB cable for some years without problems. I am using a 1k8 R in series with the PIC Rx pin, and an 18k pull-up R as recommended. In attempting to power up and use uTrace JS today, I can't get the desired response when hitting Ping from the debug window.  I have left the cable plugged in to the 1/8" TRS with uT power off many times, but I may have run out of luck. Attached are screen shots of the terminal and JS windows, which appear normal until I try to Ping the PIC. Any ideas, or should I get going on replacing the Max232 and order a new PIC?

[Ihor]

Hi Martin, 

I remember I saw some similar behavior but I do not remember who reported it and if there was a solution. In any case it would be good to know if it works with the original (Ronald's) software (the ping and acquisition) and it also would be helpful to have a similar log but not for the ping but whe you try to run a simple measurement, even for one curve with one point. The debug log should show more details about what is sent in both directions. 

[Martin Manning]

Hello Ihor, 

I tried to get the communication working using Ronald's GUI. The result was the same, no response. I haven't run the tracer in a few months, but it was working fine then. It has been sitting on my bench with the FTDI cable plugged in, but that has been the case many times before. Here is a screen capture of the terminal window after an attempted trace. Note I grabbed your latest JS release for this. Everything looks good until I click "Measure."

[Martin Manning]

Is it possible that an OS update ruined the communication? I had to buy a new FTDI cable to accommodate Windows 10. I there has been one or possibly two updates recently (currently on 14.6.1 Sonoma), but of course I am still running the legacy CPU in my MacBook, and using the x64 version of JS.

[Ihor]

I am not sure that the OS can destroy something basic so easily but who knows. To test I would suggest you to disconnect the cable from the uTracer and then connect together RX and TX pins on that adapter and try to run uTracerJS and press Measure. You can see my screenshot, one should get at least the "echo" of the command. If that does not work, there is a problem with the adapter. If that works, most likely PIC is not responding. 

I have never commented on that FTDI cable issue and precautions, probably there are some problems in some cases, but I programmed lots of PICs of different types for different projects, ATmegas, ESPs with cheapest possible Chinese USB to TTL converters (I have about 7 different types at home) and I have never had problems :) Also the cheapest USB to RS232 cables on aliexpress work perfect with my uTracer. Even connecting ESP32 (which is 3.3V device, BUT! 5V tolerable (which is hidden in the specs) to the PIC's Tx Rx directly always worked without problems. As long as the drivers for the USB to TTL adapter are properly installed (if necessary, especially for those cheap Chinese alternatives), and the ECHO works (when TX-RX are connected), everything else also worked.  

[Martin Manning]

Thanks for the reply. Here is the result with the FTDI cable connected to USB port only, Rx/Tx on the other end shorted together.

[Martin Manning]

Ok, found a problem... broken wire inside the enclosure. Repairing that now. Thanks again for the cable test procedure.

[Davo]

this is so recognizable Martin lol

The basics haha 

Ihor, 4real? I had one old chines RS232 cable, but I never made it work with the uT man, tried zo much…

It looked like this one..

But after a-lot of struggles I eventually bought an FDTI cable…

Op 28 aug 2024 om 17:46 heeft Martin Manning <mman...@fuse.net> het volgende geschreven:

Ok, found a problem... broken wire inside the enclosure. Repairing that now. Thanks again for the cable test procedure.

--

You received this message because you are subscribed to the Google Groups "uTracer" group.
To unsubscribe from this group and stop receiving emails from it, send an email to utracer+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/utracer/9193b264-4de1-4426-9238-ae6091e31cd7n%40googlegroups.com.

[Ihor]

Good to hear Martin! The screenshot looks good so the test with echo is passed so to say :) 

About the converters, that's exactly why I do not like to talk about that - for me everything and always works:) The PIC has the most standard and proper implementation of the serial communication, so if something does not work it means the problem is with the adapter and not the PIC. Now, as you can see in the photo (and it is not even everything that I could find now) I have lots of those Chinese USB cables, converters, the RS232-USB cable, Chinese RS232, and also self-made TTL-RS232 (on MAX232 or SP3232) converters which I sold together with my ESP32 module for uTracer to many users......all work with uTracer's PIC (either directly or via RS232). In the photo you can also see that PIC where I was testing my own firmware. I also connected to PIC via Arduino and ESP32 (as software simulated USB-toTTL) and all worked just fine. 

I am not saying that everything that you buy on ebay or aliexpress will work, there are lots of fakes and low quality things, so the way to go is to try. I was just lucky, 10 out of 10 so to say :)   

  

[Ihor]

I see here someone actually uses exactly the same adapter as in one of my photos ($2 or so) and build on top of it a nice extension:

https://retronics.no/2023/03/05/utracer-part-3-making-a-custom-usb-interface/

The gerber files are available here: 

https://github.com/donpivo/uTracer

So 5 of those PCBs will cost one $5 with world-wide shipping from China and then i tis a matter of soldering and having fully working USB interface :)

[Martin Manning]

Back up and running again. The break was in the Tx lead at the TRS jack. Broken, but only displaced from the solder lug 0.5mm or less. I guess the dozens of times I have opened up the enclosure over the years, each time causing those leads to flex, finally resulted in the failure. 

[Martin Manning]

Ihor wrote: I see here someone actually uses exactly the same adapter as in one of my photos ($2 or so) and build on top of it a nice extension:...

Indeed that's a very nice implementation. My solution is much more pedestrian: Remove the MAX232, add a jumper and a resistor to the empty DIP socket, and solder another resistor under the board. The existing pin header is used for the leads to the TRS jack for the FTDI TTL-USB cable. I have been using FTDI adapters since 2013, out of a desire to eliminate the bulky D9 RS232 connector. Ronald wrote it up in his web pages later, adding the resistors, where I originally just had two jumpers across the DIP socket to take the PIC Rx/Tx to the pin header.

I got to wondering if my original FTDI cable would work using MacOS and uTracer JS. Recall I had to buy a new one when I updated to Windows 10 (which I run using Bootcamp). Turns out it does! I only needed to replace the description in the uTracerJS.txt file, which was very conveniently listed for copy-pasting in the terminal window :^)

[Davo]

I’ll have to look at it tomorrow though!

It looks like I’m missing the point 

Interesting theories guy’s 👌🏻

Are there some benefits for me if it’s still working?

May e those weird errors when the HV is on 8 ur touching the software?

Op 29 aug 2024 om 03:38 heeft Martin Manning <mman...@fuse.net> het volgende geschreven:



Ihor wrote: I see here someone actually uses exactly the same adapter as in one of my photos ($2 or so) and build on top of it a nice extension:...

Indeed that's a very nice implementation. My solution is much more pedestrian: Remove the MAX232, add a jumper and a resistor to the empty DIP socket, and solder another resistor under the board. The existing pin header is used for the leads to the TRS jack for the FTDI TTL-USB cable. I have been using FTDI adapters since 2013, out of a desire to eliminate the bulky D9 RS232 connector. Ronald wrote it up in his web pages later, adding the resistors, where I originally just had two jumpers across the DIP socket to take the PIC Rx/Tx to the pin header.

<TTL_Direct.png>

I got to wondering if my original FTDI cable would work using MacOS and uTracer JS. Recall I had to buy a new one when I updated to Windows 10 (which I run using Bootcamp). Turns out it does! I only needed to replace the description in the uTracerJS.txt file, which was very conveniently listed for copy-pasting in the terminal window :^)

On Wednesday, August 28, 2024 at 3:17:15 PM UTC-4 Martin Manning wrote:

Back up and running again. The break was in the Tx lead at the TRS jack. Broken, but only displaced from the solder lug 0.5mm or less. I guess the dozens of times I have opened up the enclosure over the years, each time causing those leads to flex, finally resulted in the failure. 

On Wednesday, August 28, 2024 at 1:16:01 PM UTC-4 Ihor wrote:

I see here someone actually uses exactly the same adapter as in one of my photos ($2 or so) and build on top of it a nice extension:

https://retronics.no/2023/03/05/utracer-part-3-making-a-custom-usb-interface/

The gerber files are available here: 

https://github.com/donpivo/uTracer

So 5 of those PCBs will cost one $5 with world-wide shipping from China and then i tis a matter of soldering and having fully working USB interface :)

--

You received this message because you are subscribed to the Google Groups "uTracer" group.
To unsubscribe from this group and stop receiving emails from it, send an email to utracer+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/utracer/ec80a909-2435-4d7f-835c-766950719302n%40googlegroups.com.

<TTL_Direct.png>