RTA and Scope with high DC voltages

tonescout

Registered
Thread Starter
Joined
May 30, 2021
Messages
19
Hi,

I am hoping to use REW more with the Scope and RTA function, and I am getting confused as to how best to protect the audio interface and computer from high DC voltages along with AC signals. I have successfully used this with low level inputs but am cautious to look at more significant voltages. I had a thought that I might use a passive high impedance probe and patch this into an RCA input of my USB audio interface (https://www.behringer.com/product.html?modelCode=P0484) but others have suggested this would reduce the input signal too much. I want to see the ripple on my smoothing caps in a valve preamp at DC 680V and I want to look at the hum on signal valves that are also at 150-200V. Any ideas of best practice or if there is a 'standard' approach with this, I am also guessing a well selected capacitor would remove most of the DC, but I am hoping this is an already well trodden path?
 

DanDan

Moderator
Staff member
Joined
Mar 10, 2018
Messages
743
Buy an actual HV Scope probe and change the connector at the interface end.
 

tonescout

Registered
Thread Starter
Joined
May 30, 2021
Messages
19
Buy an actual HV Scope probe and change the connector at the interface end.
That sounds like a plan, however a friend suggested that I would get a miniscule output with an input impedance on my audio interface which is 27 Kohm input?

This is an example, hence my confusion: http://www.hantek.com/products/detail/74 it's 100Mohm, so will that give me any sort of useful signal into 27K?
 

trobbins

Member
VIP Supporter
Joined
Nov 26, 2017
Messages
121
Location
Australia
The REW forum doesn't seem like the right place for this ?

Soundcards have a very large S/N range, so 'seeing' very low signal levels along with very high signal levels is a real benefit, but you need to upgrade your awareness of how a spectrum plot tells you so much more than a scope plot as far as mains ripple is concerned.

Google the use of connecting an external capacitor to your probe, as a way of allowing AC to pass but to block DC. Connecting a 'probe' to a high DCV node requires awareness of the risks, and the input circuitry of your scope/interface - you should be able to appreciate that in some google links that show the difference in scope input circuitry that is dc versus ac coupled, and what happens if the gnd clip accidently disconnects.

Connecting a 'probe' to a high DCV node should only be done as a last resort imho. I use a lab meter that allows AC rms measurement with up to 1kVdc at the node - the rms level is a benchmark I can use, and I don't really need to see that waveform or spectrum at that node. Similarly, adding a 10:1 or 100:1 resistor divider across the B+ node, and probing the attenuated dc level is far safer as the divider is fixed in position - that can then use a standard scope input with 10:1 probe, or use a soundcard's input if ac coupled. If the DC supply is used in an amp, then the output of the amp can be probed for a spectrum and the mains ripple identified (where it counts).
 

tonescout

Registered
Thread Starter
Joined
May 30, 2021
Messages
19
The REW forum doesn't seem like the right place for this ?

Soundcards have a very large S/N range, so 'seeing' very low signal levels along with very high signal levels is a real benefit, but you need to upgrade your awareness of how a spectrum plot tells you so much more than a scope plot as far as mains ripple is concerned.

Google the use of connecting an external capacitor to your probe, as a way of allowing AC to pass but to block DC. Connecting a 'probe' to a high DCV node requires awareness of the risks, and the input circuitry of your scope/interface - you should be able to appreciate that in some google links that show the difference in scope input circuitry that is dc versus ac coupled, and what happens if the gnd clip accidently disconnects.

Connecting a 'probe' to a high DCV node should only be done as a last resort imho. I use a lab meter that allows AC rms measurement with up to 1kVdc at the node - the rms level is a benchmark I can use, and I don't really need to see that waveform or spectrum at that node. Similarly, adding a 10:1 or 100:1 resistor divider across the B+ node, and probing the attenuated dc level is far safer as the divider is fixed in position - that can then use a standard scope input with 10:1 probe, or use a soundcard's input if ac coupled. If the DC supply is used in an amp, then the output of the amp can be probed for a spectrum and the mains ripple identified (where it counts).

Thanks for some good guidance, I will do some reading over the weekend. It maybe the case that saving money on a conventional oscilloscope and spectrum analyser comes with too many risks :-)
 
Top Bottom