BK Precision 1735A Power Supply - Display fluctuating up and down

[cman]

I have a BK Precision 1735A Linear Power Supply - It has two displays. Each display is on its own little daughter board. One for the Voltage which is the green one on the top. And then another for current, which is the red one on the bottom.

Sometimes the voltage readout on the top just randomly fluctuates up and down to completely random numbers.. its never an illegible random segment, its always an actual number on the 7 (or 8?) segment display.. but its just fluctuating all over the place. This is intermittent and only happens sometimes. The actual voltage output of the power supply never changes unless i move the knobs.. so that is great.. but I would like to get the voltage readout fixed.


The current display on the bottom works just fine.

I have uploaded some pictures and a video. I see a quartz crystal on the board labeled "100000" (1 with 5 zeros) - The only thing I could find on Digi-key for sale was 10mhz and 100khz which had similar numbers on them... can someone help me identify what model or frequency this crystal is, and also give some insight as to if you think this could be the problem?


The reason I believe it is the crystal oscillator is because when the problem is not happening (its intermittent) and I touch the crystal the problem returns... i think it may be temperature or something causing the crystal to drift, etc.




Video I uploaded of myself touching the crystal to recreate the problem (or so i think) (video is processing should be able to be viewed in about 10-15 min)

https://drive.google.com/file/d/1FaMwB28XaU89puKr8M0ndzrEXttKLGhS/view?usp=sharing


Pictures in post below

[cman]

The rest of the pics here.

[bdunham7]

AFAIK there is no service manual or schematic available, so a bit of reverse engineering or similar thinking is required.  Fortunately it is a pretty straightforward here.  Yes, a malfunctioning crystal (this is a 100kHz tuning fork crystal) could be the cause and touching, tapping, cooling or heating crystals can cause them to start or stop working.  A crude but effective diagnostic approach!

The ADC chip is the TLC1735C and the datasheet is below.   If you read it, you'll see that a 10MHz crystal is not possible, but 100kHz is a likely choice.

https://www.ti.com/lit/ds/symlink/tlc7135.pdf?HQS=dis-mous-null-mousermode-dsf-pf-null-wwe&ts=1656179448241&ref_url=https%253A%252F%252Fwww.mouser.com%252F

And while selecting the crystal would seem to require looking at the other chips and their datasheets (the ADC appears to need an external clock circuit, not just a crystal), I'd bet $2 that this one will work:

https://www.mouser.com/datasheet/2/122/ecs-31x-669.pdf

[MikeK]

That's not a dirt cheap power supply.  I would contact BK, explain the situation, and ask for help...possibly a schematic.  They might be familar with the problem.

[cman]

That's not a dirt cheap power supply.  I would contact BK, explain the situation, and ask for help...possibly a schematic.  They might be familar with the problem.

Yes, definitely not cheap.  I got it "As-Is" on ebay for under $100. I knew that it might have issues, but I figured it would be fun to fix and save several hundred dollars. I have reached out to BK, and they have offered up a replacement board for $75 USD plus shipping. Which is not cheap, but I believe not unfair in any way considering its the whole board they would be replacing. So this is kind of my "Plan A" to fix it and then buying that whole replacement board is "Plan B"

[cman]

AFAIK there is no service manual or schematic available, so a bit of reverse engineering or similar thinking is required.  Fortunately it is a pretty straightforward here.  Yes, a malfunctioning crystal (this is a 100kHz tuning fork crystal) could be the cause and touching, tapping, cooling or heating crystals can cause them to start or stop working.  A crude but effective diagnostic approach!

The ADC chip is the TLC1735C and the datasheet is below.   If you read it, you'll see that a 10MHz crystal is not possible, but 100kHz is a likely choice.

https://www.ti.com/lit/ds/symlink/tlc7135.pdf?HQS=dis-mous-null-mousermode-dsf-pf-null-wwe&ts=1656179448241&ref_url=https%253A%252F%252Fwww.mouser.com%252F

And while selecting the crystal would seem to require looking at the other chips and their datasheets (the ADC appears to need an external clock circuit, not just a crystal), I'd bet $2 that this one will work:

https://www.mouser.com/datasheet/2/122/ecs-31x-669.pdf

Great, thank you so much for this information. I was thinking the number of zero's matched the 100khz crystal, I was just not entirely sure as this is my first time dealing with a crystal and this type of circuit. It looks like they are very cheap, I am going to grab one and try it out, the worst case is I will lose $5 shipping and a few bucks.


I got this for very cheap knowing it may be broken.. but I am a college student going to school for electrical engineering so I am trying to get equipment for good prices, and it gives me some good circuit practice working on broken items as well.

[bdunham7]

That's not a dirt cheap power supply.

Indeed, unlike many others--even some rather expensive A-brand models--these won't blow up if you try and charge a battery, even a car battery and even if you accidentally hook it up backwards.  I'm fairly sure BK Precision doesn't have the schematic for the display boards, although I suppose it wouldn't hurt to ask.

[mawyatt]

If you have access to a AWG or other source, set the source output peak to peak squarewave to ~ expected supply voltage, say 3VPP. Then prepare a simple probe of a series R of 1~10K and a series C of 1~10nF. Connect the probe to the source use the probe to excited either wire of the crystal or other components that might have something to do with the clock generation, and adjust the frequency of the source to the expected frequency of the clock. This might help isolate the problem and identify the proper frequency if the crystal is the issue.

Hope this helps and good luck,

Best,

[MikeK]

You can probe the clock signal at pin 22 of the ADC, and there's a convenient test point via there.  Looks like U3 is part of the clock generator circuit?

[cman]

AFAIK there is no service manual or schematic available, so a bit of reverse engineering or similar thinking is required.  Fortunately it is a pretty straightforward here.  Yes, a malfunctioning crystal (this is a 100kHz tuning fork crystal) could be the cause and touching, tapping, cooling or heating crystals can cause them to start or stop working.  A crude but effective diagnostic approach!

The ADC chip is the TLC1735C and the datasheet is below.   If you read it, you'll see that a 10MHz crystal is not possible, but 100kHz is a likely choice.

https://www.ti.com/lit/ds/symlink/tlc7135.pdf?HQS=dis-mous-null-mousermode-dsf-pf-null-wwe&ts=1656179448241&ref_url=https%253A%252F%252Fwww.mouser.com%252F

And while selecting the crystal would seem to require looking at the other chips and their datasheets (the ADC appears to need an external clock circuit, not just a crystal), I'd bet $2 that this one will work:

https://www.mouser.com/datasheet/2/122/ecs-31x-669.pdf

She is fixed!! It was the 100khz crystal. It cost me $0.75 from digikey!


Thank you for the help everyone!