16-Bit PSN-ADC-USB, Crystal and Scale Issues

[carlbe...@hotmail.co.uk]

Hello All.

I don't know if anyone can help, but I recently finished building the 16-Bit PSN-ADC-USB and I'm having some trouble adjusting the crystal and Scale.
 
Crystal: In the build_info it says to use a 74HC14 Hex Schmitt-trigger to measure the frequency, which i have done. When adjusting C27 (9-50pf ceramic trimmer) nothing happens, it just jumps from 7.98Mhz to 8.02Mhz regardless if i alter the pot or not. When measuring the frequency without the 74HC14, just measuring pin 14 i get the same.
 
Scale:  When the board is connected to my PC running WinSDR, looking at the ADC counts, providing 10v into pin 1 with the jumper shortened and adjusting Scale, again nothing happens. The ADC counts stay the same.
 
Could i have flashed the wrong Hex data? In the hexfiles i uploaded 'SdrBoot3014_Usb.hex' and using command prompt ran 'dsdr26_usb.hex', both where successful.

I have checked all my solder joints and everything looks good to me. When powering the board for the first time there was no smoke or heating of the IC's.

Also, when tapping the inputs with my finger i see small jumps/spikes on the graph on WinSDR and i have no issue adjusting the Offset.

Thank You.

[Larry Cochrane]

See comments below...

On 4/22/2026 1:34 PM, carlbe...@hotmail.co.uk wrote:
> Hello All.
>
> I don't know if anyone can help, but I recently finished building the 16-Bit PSN-ADC-USB and I'm having some trouble
> adjusting the crystal and Scale.
>

> _Crystal_: In the build_info it says to use a 74HC14 Hex Schmitt-trigger to measure the frequency, which i have done.

> When adjusting C27 (9-50pf ceramic trimmer) nothing happens, it just jumps from 7.98Mhz to 8.02Mhz regardless if i alter
> the pot or not. When measuring the frequency without the 74HC14, just measuring pin 14 i get the same.

You might be overloading the crystal. Adding a FET amplifier/buffer, with low input capacitance, instead of 74HC14
between the crystal and the frequency counter may stabilize things.

>
> _Scale_:  When the board is connected to my PC running WinSDR, looking at the ADC counts, providing 10v into pin 1 with

> the jumper shortened and adjusting Scale, again nothing happens. The ADC counts stay the same.

I ran into this problem now and then when I was producing the board. To fix the problem I would lower the 1 meg resistor
going to the scale pot to around 800k to 900k.

Regards,
Larry Cochrane
Redwood City, PSN

[Carl]

Hi  Larry,

Thank You for your reply.

I will give this ago, I just need to order a  FET amplifier. I'm going to use a MPF102 and will report back here if it worked or not.

On the seismicnet site it shows C52 (68uf) as unpopulated, is this correct?

Thanks Again for the help.

[Larry Cochrane]

Hi Carl,

> I will give this ago, I just need to order a FET amplifier. I'm going to use a MPF102 and will report back here if it
> worked or not.

When you make the measurement be sure to connect the amp/counter to the crystal output driver on the DspPic IC. The
effect of the extra capacitance is less if you use the crystal driver output not the crystal input pins.

>
> On the seismicnet site it shows C52 (68uf) as unpopulated, is this correct?

I'm leaving town for a few days so I don't have time to look this up. What page/document do you see this note on?

-Larry

[Carl]

C52. It was just in the picture (https://www.seismicnet.com/serialatod.html) and i dont see it in the BOM, so i assumed it was not populated.

[Carl]

I have tried different resistors (680k, 750k, 820k, and 910k) and non made any change. During each test i applied 10v into input 1 and had the jumper closed. Adjusting the Scale all the way in each direction made no change.

I also tried the MPF102 to measure the crystal and nothing, i still get the same thing. I removed the ceramic cap to see if it was damaged and it works perfectly. Also, as suggested by Google Gemini removed C28, it though that the capacitance may be to high. It did nothing. 

I'm stumped as what to do next. I don't think any of the IC's are damaged, as adjusting 'Offset' works and when i tap the inputs with my finger i see change on the graph on WinSDR. Also, when i apply 10v the graph shoots up and i get an alarm.

[Carl]

I have rebuilt the board as i thought there might be some broken traces.

And i have been going back an forth with Google Gemini. It thinks that there is an issue with my power, but i don't trust what it says.

Are you able to tell me what the voltages are for (U8)LTC1605 pin 1, 10 and (U7)DG508 pin 12.

Thank You.

[Larry Cochrane]

One thing you need to double check is that you have the right version of the LTC1605. Linear Tech makes 3 versions of
the chip. The LTC1605 has a +-10V input range, the LtC1605-1 has a 0 to 4v range and the -2 has a +-4v range. The board
and software was designed for the the LTC1605 part not the -1 or -2 parts.

On 5/7/2026 12:48 PM, Carl wrote:
> I have rebuilt the board as i thought there might be some broken traces.
> And i have been going back an forth with Google Gemini. It thinks that there is an issue with my power,

You should see a steady +5V and +- ~13 volts from the LM2574 switching regulators.

>but i don't trust what it says.

I'm not sure how AI will help you debug a board that it knows nothing about, but what do I know...

> Are you able to tell me what the voltages are for (U8)LTC1605 pin 1

Don't know. Because of the 200 and 33.2k ohm resistors it will be a somewhat lower then the input voltage.

>Pin 10

This is one of the digital output pins. It is not connected to the DspPic. The cpu only reads the lower 8 bits from the
ADC and toggles the BYTE pin to read the other 8 bits.

and (U7)DG508 pin 12.

This is channel 5 input. If you have nothing connected to channel 5 you should see 0 vdc.

-Larry

[Carl]

Hello, Thank You for your reply.

I rebuild the board and all my voltages are stable, I'm using a 15v power supply.

I still cannot adjust the Scale, nothing happens when i adjust the pot. I have attached a two screen shots of WinSDR showing what happens when i connect 10v and disconnect it. Maybe it helps.

I believe the LCT is the correct one, iv attached a photo hopefully you can see it. 

I gave Gemini the schematics, BOM and build info. So it new what the board was about. It did also mention the variants of the LTC device.

I have been trying the MPF102 but don't see any difference on my scope, i get the same readings. Perhaps i have it set up wrong? I connected my probe to 'Source' pin.

 

[Carl]

Finally successes.

I don't know why it worked but here is what i did. 

Apply 10v to pin 1 before plugging USB into the board, that's it.

I would normally have the board powered and plugged into my PC with WinSDR opened, then apply 10v.

But this time i did it before and now i can adjust the Scale.

Also, i can see and adjust the crystal. This was my error i had gate and source mixed up.

Thank You all for your help and sorry if i was a pain. 

[Larry Cochrane]

Hi Carl,

It looks like you did a good job at building the board. I'm also impressed that Gemini would recommend you check that
you have the right LCT part. From the picture it look like you have the correct ADC version.

Here is a test I suggest you try. Temporarily remove one end of the 200 ohm resistor, the one going to the LF411, from
the PCB. This will remove the LF411 op-amp and DB508 mux chip from the ADC input. Now ground the free end of the 200 ohm
resistor using a clip lead. Startup WinSDR and make sure you have it set to record one channel. You should see a
continuous straight line on the screen with the average ADC count around zero +- a few counts. If you get a straight
line near zero you should be able to use the offset pot to set to zero.

Let me know how this test goes. One more thing. Before the test delete the current winsdr.log file and start the test.
After the test post it here so we can take a look at it. Maybe something in it will help debug the problem.

-Larry

On 5/23/2026 9:34 AM, Carl wrote:
> Hello, Thank You for your reply.
>
> I rebuild the board and all my voltages are stable, I'm using a 15v power supply.
> I still cannot adjust the Scale, nothing happens when i adjust the pot. I have attached a two screen shots of WinSDR
> showing what happens when i connect 10v and disconnect it. Maybe it helps.
> I believe the LCT is the correct one, iv attached a photo hopefully you can see it.
> I gave Gemini the schematics, BOM and build info. So it new what the board was about. It did also mention the variants
> of the LTC device.
> I have been trying the MPF102 but don't see any difference on my scope, i get the same readings. Perhaps i have it set
> up wrong? I connected my probe to 'Source' pin.
>

> 20260523_171057.jpg
>
> 10v Connected.png 10v disconnected.png

[Ted Blank]

Carl, many years ago I spent many months trying to understand why my home built amplifier board wasn’t working. I was trying to get both + and -12 V out of two  6V batteries wired in series center tapped.   Of course you need four batteries in series for that.   It’s a lesson you never forget. 

Ted

--
You received this message because you are subscribed to the Google Groups "Public Seismic Network Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to psnlist+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/psnlist/bd74e94a-24af-4745-812d-1a504484d1c1n%40googlegroups.com.

[Jonathan]

Carl,

Was this a blank board? The gerber files that Larry put on his GitHub for this board had the SOIC footprint of the ADC and I believe was serial only. I ordered some from JLC and have been meaning to build them along with SOIC ADCs. 

He changed it because only the SOIC package was available and the DIP was discontinued.

Jonathan

To view this discussion visit https://groups.google.com/d/msgid/psnlist/CAC8W%3Dt2%2B19GP-N2g_ZsPHuCeh9EszUy7PaAokNYQMe5pKKd4cA%40mail.gmail.com.