Probing around valve amps with an Oscilloscope

[g1]

If it's AC coupled and you put 500V DC on the input the coupling capacitor charges up to 500V through the 9M and 1M resistors as in the sketch below.

Thanks for clearing that up. I had assumed the coupling cap was after the scopes input resistance but you are correct.
So on AC coupling, even with a 10:1 probe we can NOT exceed the DC voltage rating of the scopes coupling cap. The scopes cap will see the full DC present at the probe tip.

[JoeM]

Thanks. I do see how it could go either way. But it seems to me that it would be clearer to call it a 10th probe, or a /10 probe. Just sayin'. 10X, to my mind, implies "10 times". And that the opposite of what the probe does. I suppose it's all good as long as the implemented nomenclature is understood. But with translation of languages, as well as typically understood grammar, it seems that the standard designation for scope probes is less than ideal. We need a scope probe designation revolution!

Yes, but with a 10x probe the circuit is loaded less than the typical 1meg input of the scope by a factor of 10.

[JoeM]

If it's AC coupled and you put 500V DC on the input the coupling capacitor charges up to 500V through the 9M and 1M resistors as in the sketch below. I think the capacitor in my scope is only rated at 400V. If I want to check the ripple on the B+ I set the scope to DC and use a 10n high voltage capacitor clipped in the probe.
[ATTACH=CONFIG]21326[/ATTACH]

Dave, I always thought the 500v limit has to do with the scope probes compensation capacitor used to give a flat square wave- you don't show that in your schematic.

[Steve Conner]

If this is the case, how can they make x100 probes with 100M input impedance? I have a pair that work fine.

[Dave H]

Dave, I always thought the 500v limit has to do with the scope probes compensation capacitor used to give a flat square wave- you don't show that in your schematic.

To keep it simple I omitted the probe compensation capacitor and the 1M input resisor is really the scope's 1M compensated attenuator. That's how my homemade scope is anyway. I'm not sure all commercial scopes are the same.

[Tom Phillips]

There are a couple of things to consider when you are measuring high voltages with your scope. The scope and the probe work together as a system. Each has their own maximum voltage specification. If you are using a scope and probe made by the same manufacturer and intended to be used together then the specifications given in the probe manual will be the overall limits. Otherwise, you may need to consider the individual probe and scope specifications and adhere to the one with the lower limit.

As Dave H pointed out in post #15 If you use AC coupling on your scope input, then, even if you are using a 10X probe your input coupling cap will charge to the full DC component of the signal being measured. The Maximum input voltage spec for my Tektronix model 465 scope is:
DC Coupled “250 V (dc + peak ac) or 500V p-p ac”
AC Coupled “500 V (dc + peak ac) or 500V p-p ac”
By the way, the input cap used in this scope is a 0.019μF/ 600V rated part.

Given these specs, what are we to do when you want to look at the ripple riding on a 650V B+ supply or the drive signal at the output transformer plate winding. There is a thread on MEF somewhere discussing the dangers of scoping the OT plate winding and the safe method proposed was to use a 100X probe. The answer is the same when you want to look at other HV signals with either DC or AC coupling to your scope’s vertical amplifier. The reason a 100X probe can solve the problem is because many 100X probes utilize a different circuit design than the common 10X probe. As Dave H showed, a 10X probe basically just inserts a 9 MΩ series resistor in the signal path. Reference the schematic for the Tektronix model P6007 100X probe (attached) and you will see that there is a resistive divider made up of a 9.9 MΩ and a 111 kΩ resistor. With this design the signal presented to the input of the scope’s vertical amplifier is attenuated by ~100 from DC to the frequency limits of the probe.

Remember that the scope and the probe working together from a system. With my scope and a common model P6105 10X probe the max voltage spec from the probe manual is “500 V dc + peak ac” derated with frequency. Note that using the 10X probe did not raise the max voltage limits given in the scope specifications. The derating doesn’t start until > 1.5 MHz so this is not a concern for the guitar amp service tech. I just included it here for completeness and so that readers are aware that there are additional factors that come into play when you have the need to do more specialized work and need to make the most precise measurements.

Now let’s discuss the max voltage spec that we can expect with a 100X probe. We need to refer to the probe manual. I can’t say if all manufacturers use similar designs but in the case of the P6007 probe I listed above the maximum voltage specification is given as “1.5 kV dc or rms (2.1 kV peak or 4.2 kV peak-to-peak)” Again there is a voltage vs. frequency derating curve which shows that voltage reduction begins at 200 kHz. Note that at 1.5 kV dc there is only about 15 V dc at the scope input. Since we know that the scope input will withstand a much higher voltage, this tells us that the voltage limits are set by the probe design. For those who have the need there are probes available with much higher limits. (Just be prepared to pay more for the performance)

The P6007 100X probe is a relatively simple design. Its rated band width is DC to 25 MHz. It has been out of production for many years. However, with patience and research good serviceable units can be had for less than the cost of a pair of current production 6L6s. There are, of course, many other probe models available from several manufacturers. You just need to check the specifications to verify that the probe is suitable for your needs. Some 100X probes have very low max voltage specs (less than 50V)

I’ve been paying more attention to these details lately. I’m a test equipment freak and, unless there is a disaster, I probably have a lifetime supply of bench equipment and backup spares for my own use. Over the years I have damaged a few pieces of test equipment with improper procedures and general carlessness. I’m doing my best to be more careful these days. It takes enough time as it is to maintain my vintage test equipment. I don’t want to add damage repair and the attendant search for obsolete components to the workload.

I use a /10 probe (didn't use an "X" because it's a division, not a multiple). But I guess, for whatever reason, it's called a 10X probe!?! ...

Chuck,
The 10X, 100X convention goes back to the days before scopes had automatic vertical scale readouts. The convention was adopted to mark the probe with the multiplication factor that should be applied to convert the reading to the correct value for the probe being used.

I hope you find this information helpful,
Tom

[Tom Phillips]

Attached is the Tektronix P6007 scope probe schematic that went missing from post #21 of this thread. I thought it would be useful to have since the thread has been linked in other discussions and several people seem to be interested in understanding how to safely probe high voltage signals.
Cheers,
Tom

[Chuck H]

Since this thread has resurfaced I'd like to clear something up... I said:

Thanks. I do see how it could go either way. But it seems to me that it would be clearer to call it a 10th probe, or a /10 probe. Just sayin'. 10X, to my mind, implies "10 times". And that the opposite of what the probe does. I suppose it's all good as long as the implemented nomenclature is understood. But with translation of languages, as well as typically understood grammar, it seems that the standard designation for scope probes is less than ideal. We need a scope probe designation revolution!

The responses have been clear and I understand the logic of the markings, but...

To my mind it would still be better to mark MODERN probes to indicate the PROBES function rather than the scopes function when used with that probe. "My" logic gets skewed when I have to interpret the individual components markings and function (the probe) based on it's interaction with a whole system. Perhaps this is also confusing to new scope owners that are trying to choose probes. At the very least it requires additional explanation when telling a new scope owner about the scope/probe relationship. If I'm thinking about the probe, like reading the markings on the probe, I want to know about the probe. I can determine it's effect on the system for myself from there. So I maintain that the current system is unnecessarily confusing. But I digress. I don't need any further excuses or endorsements for the existing system. I get it.

[nsubulysses]

great thread