0
\$\begingroup\$

CH2 of my Rigol DS1054Z is bad, it fails calibration even after the scope is warmed up, amplitude of the captured signal is lower than other channels, DC bias is higher than other channels.
I have already followed Rigol debug steps and no factory reset, no FW reflashing solves it. They think it is a hardware issue, but the scope is out of warranty.
CH1 is good, I know it passes calibration and the captured amplitude is correct. I have these reverse engineered schematics of the front end: front end 1 front end 2

I soldered some wires and I captured the 0/3V reference square wave into CH1 and CH2.
The output of the differential amplifier seems wrong, the amplitude of CH2 is much lower than CH1: diff amp out probe CH1 diff amp out CH2 diff amp out

I also debugged a few other points of the circuit. I probed the output of the "FET Input Amp" to the differential amplifier, and this is what I got: out to diff amp CH1 fet amp out CH2 fet amp out

This lead me to believe there's some kind of problem with the DC gain/vertical offset components, because of this DC bias difference.
So I probed pins 1, 2, 3 of the TLV274.
These are for pin 1: probe tlv274 pins 1,2,3 ch1 tlv274 pin 1 ch2 tlv274 pin 1

These are for pin 2: ch1 tlv274 pin2 ch2 tlv274 pin2

These are for pin 3: ch1 tlv274 pin3 ch2 tlv274 pin3

I initially thought this meant the TLV274 was the culprit, because its output (pin 3) has a considerable difference between CH1 and CH2.
However, I now think this might be misleading, because the schematic omits feedback from the FET amplifier output through resistor 806K going to the inverting pin 2 of the Op-Amp.
Dave Jones specifically said there must be feedback, however he could not find it.
Just to confirm it, I probed pin 10 of the TLV274, and this is what I got: tlv pin 10 probe CH1 tlv274 pin 10 CH2 tlv274 pin 10

I have the following questions:

  1. Is there a way to eliminate the effect of the TLV274 output into the "FET input amp" stage, so I can confirm or rule out the TLV274 as being the culprit? Could I just remove the 4.7Meg resistor going to the JFET gate and measure the FET amplifier output?
  2. Are there other easy tests I could do? I am checking the power rails at the moment, and I plan to desolder the BAV199 clamping diodes to confirm they are not biasing the signal.
  3. Have I missed anything during debug, and the waveforms indicate the problem come from elsewhere?
\$\endgroup\$
4
  • \$\begingroup\$ I wonder if you would be better on eevblog? He has a forum. It's where I got my DS1054 information from. \$\endgroup\$ Commented May 10 at 9:10
  • \$\begingroup\$ @Neil_UK that has already been done, but the topic didn't get much attention. Maybe I will try to send a message to the DS1054Z main topic to see if someone is willing to help. \$\endgroup\$ Commented May 10 at 10:36
  • \$\begingroup\$ I would do much more of the testing with just a meter and a DC input, your problem exists at DC and it is confusing to have so many measurements shown. Add reference designators to the schematic - we can't easily refer to the circuit otherwise. Measure what happens when you adjust the vertical position for example. The TLV274 section 1 is there to reconstitute the DC component of the signal while the other parts are to add the vertical position offset into the feedback. The overall gain both AC and DC to the output of the input amplifier MUST be unity (maybe with an offset). \$\endgroup\$ Commented May 10 at 13:48
  • \$\begingroup\$ @KevinWhite I plan to do a proper schematic with some inconsistencies I found while debugging in the near future. So about your suggestion to inject DC, I fail to understand how it will help with my problem. As I said, my initial assumption was the TLV274 or the passive components surrounding it must be faulty. However, there's this feedback from the output of the FET amplifier that goes back into the TLV274 pin 2, not depicted in the schematic. How do I rule out a DC bias being created by faulty components on the FET amplifier path? This is my question 1 from above. \$\endgroup\$ Commented May 10 at 16:48

0

Reset to default

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.