Oscilloscope Channel 1 Input Issue

Thread Starter

Melvo

Joined Feb 19, 2018
19
Hello All,

One of the Oscilloscopes that I am using has an issue with the Channel 1 input. When nothing is connected to the scope probe on Channel 1 the trace starts to ramp up very slowly indefinitely. When the scope probe is connected to a circuit there is an offset voltage of ~100mV. I have confirmed that there is not actually any voltage on the probes to cause this offset with a calibrated multi-meter and by using another channel on the scope.

Another weird thing is that when I connect my USB Drive with a metallic housing I see a charge/discharge to the ground on the scope trace. When I connect my USB Drive with a plastic housing nothing happens.

Here is the 100mV offset voltage with the probe connected to a circuit/device but no actual voltage there.
100mV Offset Channel 1.png

Here is the Trace with a slight drift. It is a little tough to tell completely because of the small time base.
Drift.png

This is what happens when I connect/disconnect the USB with a metallic housing.
Metallic USB Connection.png USB Connection.png

Here is what the connection/disconnection looks like with a larger time-base.
Drift and USB Connection.png

Could this issue be coming from an overvoltage condition on the scope channel? What component of the input circuitry has been damaged, if any? No other channel has this issue and the scope can handle 300Vrms.
 

AlbertHall

Joined Jun 4, 2014
12,345
'Scope inputs generally use FETs with protection diodes to ±supply. It sounds like one of the protection diodes and/or FETs is leaky.
This might happen spontaneously or by input overload.
 

Thread Starter

Melvo

Joined Feb 19, 2018
19
'Scope inputs generally use FETs with protection diodes to ±supply. It sounds like one of the protection diodes and/or FETs is leaky.
This might happen spontaneously or by input overload.
Albert,

Thanks for your reply! I can't seem to find the input schematic for my particular scope. Might you have a generic schematic that illustrates this concept? I know that most scopes have a 1Mohm resistor in parallel with a ~20pF capacitor but I am not sure what is beyond that.

I will not be working on the scope I mentioned myself since it is a work scope I just want to understand the problem so that I can fix it if it ever happens to my personal scope.

Thanks,
Melvin
 

Wuerstchenhund

Joined Aug 31, 2017
189
Could this issue be coming from an overvoltage condition on the scope channel? What component of the input circuitry has been damaged, if any? No other channel has this issue and the scope can handle 300Vrms.
Doing remote diagnostics is difficult but it could be an issue with ADC BIAS, either with the voltage supply or one of the ADCs.

Since the MSO6104A has four 2GSa/s ADC hybrids which in 1ch and 2ch mode are used in interleaved mode to get 4GSa/s, it would help to know if the problem appears in interleaved mode (i.e Ch 1 on, all other channels off) and in non-interleaved mode (i.e. all channels on).

If it's a voltage supply issue then it should be fixable, however if it's an ADC issue it will have to go to Keysight for repair.
 

Thread Starter

Melvo

Joined Feb 19, 2018
19
'Scope inputs generally use FETs with protection diodes to ±supply. It sounds like one of the protection diodes and/or FETs is leaky.
This might happen spontaneously or by input overload.
Albert,

I see. This is very complex circuitry but I think I understand the reason why this could be an issue.

Both D301 and D302 should be in reverse bias under normal operation but this small leakage current could be flowing onto the gate of TR301. Causing this transistor to go into its active region. This in turn triggers TR302 to activate the differential pair TR303 that goes to the TRIGGER TO PCB.

At least that is what I can gather from that portion of the schematic, I wouldn't be surprised if I was way off lol.

Thanks
 

Thread Starter

Melvo

Joined Feb 19, 2018
19
Doing remote diagnostics is difficult but it could be an issue with ADC BIAS, either with the voltage supply or one of the ADCs.

Since the MSO6104A has four 2GSa/s ADC hybrids which in 1ch and 2ch mode are used in interleaved mode to get 4GSa/s, it would help to know if the problem appears in interleaved mode (i.e Ch 1 on, all other channels off) and in non-interleaved mode (i.e. all channels on).

If it's a voltage supply issue then it should be fixable, however if it's an ADC issue it will have to go to Keysight for repair.
Wuerstchenhund,

I never thought about that. I only tried with the USB in the drive and out of the drive and noticed that the trace is brought up/down to ground once the USB is inserted. I will look into this and get back with you on Monday.

I probably won't fix the issue myself in any case because it is a calibrated scope. I just want to understand the nature of the problem and make note of it just in case I ever see it again down the line elsewhere I will at least have somewhere to start from. I will also let you guys know what the issue was once it comes back from repair, but before I send it out for repair I will check the difference in performance from interleaved to non-interleaved mode.

Thanks
 

Wuerstchenhund

Joined Aug 31, 2017
189
I probably won't fix the issue myself in any case because it is a calibrated scope. I just want to understand the nature of the problem and make note of it just in case I ever see it again down the line elsewhere I will at least have somewhere to start from. I will also let you guys know what the issue was once it comes back from repair, but before I send it out for repair I will check the difference in performance from interleaved to non-interleaved mode.
Good idea.

Keysight will probably just replace the main PCB as they usually do with these scopes.

Anyways, once it's fixed, if you plan to use it for a few more years then I'd recommend to consider buying a Keysight Repair Agreement for the scope which is pretty much factory warranty which can be bought for HPAK equipment even if the original warranty has expired, as long as the instrument is working (which it will be after repair) and still supported (which the MSO6kA is). And quite often Repair Agreements are surprisingly cheap.
 
Top