Patch Cables for audio

[adamboon]

Hi,
So we've got a Rigol DS1054Z and a Siglent SDG1025 to work on some Audio things with.  Now these units both have BNC's on the front panel's.  I know test gear likes to work at 50 ohms impedance, but I'm not sure how to integrate it with our world.  So the reference impedance for Balanced Audio is 600 ohms, and the output impedance is usually lower (down to around 100 ohms) and the input impedance higher (usually 600 ohms - 10k) which means the signal gets driven from the output to the input.

I believe the 1054z can be "Switched" to terminate the input to 50 ohms, but otherwise if very high (1 million ohms or so).  and I believe the signal generators get a few more volts (unto double) available to impedances above 50 ohms compared to driving 50 ohms.

so everything should be fine?

but what cable to I use (or doesn't it matter)?  should I just buy some thin 50 ohm coax leads and put XLR's (or TRS Jacks) where required?  does it matter if it's 50 ohm or 75 ohm?

And as both the signal gen and scope are "Unbalanced" I assume I short pins 1 & 3 (the inverted and the shield)?

and if I wish to terminate the input of the scope to 600 ohms, i'd just place a 600 ohm resistor across 2 & 3 (Non-Inverted and Inverted)?

I also will need to have a meter in parallel with my scope inputs sometimes - BNC "T" piece to split it? BNC -> Banana plugs into the meter?

Our HP Audio Sine generator has Banana plugs out and a pair that says 600 ohms.  How do they work (they're the ones I've always used).  they look like their floating from ground (there is also a ground point) so do I wire that as though it's balanced? (shield to shield, 2 to hot, 3 to cold).

[Rerouter]

most audio gear is low pass filtered to a few hundred Khz at best, so your not really getting into the controlled impedance concerns of a 20 MHz or higher scope, you treat it more like an ADC input, where you care more about good low noise return paths for the signal (grounds) and excluding external noise,

The rated output load of an amplifier is where it will give its rated output amplitude, for such low frequencies you can place a resistor that matches its rating at either end of the cable and measure with the score in high impedance,

[alsetalokin4017]

No, the Rigol DS1054z does _NOT_ have switchable input impedances. The channel input impedances are fixed at 1 Megohm. If you really need 50 ohm input impedance for some reason you will need to supply a 50 ohm adapter, either a feedthrough type or a BNC T + 50 ohm terminator, at the scope channel's input BNC connector.

https://www.amazon.com/Rigol-ADP0150BNC-Ohm-Impedance-Adapter/dp/B01C3EFUF0

You should definitely check your Sine Generator to make sure that the outputs are floating. Many signal/function generators with BNC output connectors are _not_ floating but rather they have the BNC shield connected to chassis ground, which in turn will be connected to the mains cord ground pin. The SDG1025 is probably like this, with BNC shields, hence the "black" or "-" output lead, grounded. Since you have banana connectors and a separate ground connector on the HP sine generator, you are probably correct that your outputs of this unit are floating, but you should definitely check to make sure, because the oscilloscope's BNC shields (and hence the Probe ground reference clips) are grounded back to the mains cord ground pin, and you don't want to be creating any inadvertent groundloops. This goes double for when you are scoping around inside amplifiers, especially tube types. Don't inadvertently ground some part of your DUT by improper scope or signal generator connections! Definitely do research how to make differential measurements using two scope channels.

[uncle_bob]

Hi

Ok, lets back of a bit:

What do you actually want to do? "Play with audio" is pretty broad.

If you want to work with balanced inputs and outputs you will need a way to properly convert from un-blanced to balanced. That's as easy as a transformer in some cases. In others op-amps are a better choice. You can do it for a few dollars or a few thousand dollars. I would suggest that the "few dollars" approach makes more sense with the gear you have.

If you are doing troubleshooting, a scope with a proper probe is a useful tool without an strange add ons. A lot of audio gear has un-balanced inputs. Driving them with a normal function generator is simply a matter of wiring a cable with the right connectors on both ends.

Doing high precision measurements at audio takes gear that is designed to do that sort of stuff. What you have are a pair of great general purpose instruments. I would not go to overboard adapting them for this and that. Use them as what they are. If you need to get fancy, look into gear that is specific to the task you want to accomplish.

Bob

[bson]

You don't need to impedance match audio signals, or worry about cabling. You can just clip onto an XLR jack directly using 10:1 passive probes, one for each polarity, and use math functions on the scope to measure various aspects of the differential signal that you're interested in.  Unless you have tremendously long cabling the impedance doesn't matter for signal fidelity.  However, you might want to use different load resistors representing an actual common load range, such as 20k-50k to verify the source doesn't care.

[f5r5e5d]

scopes, function generators are invaluable for general debugging, getting the circuit to basically work, evaluate stability at the typical low MHz unity gain intercept frequencies

but eyeballs on scope traces aren't anywhere near the resolution you want once the circuit is working - maybe few % distortion is visible on a sine wave trace

PC sound cards, even motherboard sound chipsets are orders of magnitude better for distortion, frequency, phase response, collecting high res data for discrete fft at conventional audio frequencies

[macboy]

scopes, function generators are invaluable for general debugging, getting the circuit to basically work, evaluate stability at the typical low MHz unity gain intercept frequencies

but eyeballs on scope traces aren't anywhere near the resolution you want once the circuit is working - maybe few % distortion is visible on a sine wave trace

PC sound cards, even motherboard sound chipsets are orders of magnitude better for distortion, frequency, phase response, collecting high res data for discrete fft at conventional audio frequencies

This is very true.
I recently bought a M-Audio Audiophile 192 for exactly this purpose.

[adamboon]

I'm initially using this stuff to (most of all) ensure things aren't oscillating (at Higher than Audio Frequencies), and aligning the Azimuth of tape head stacks.

Have made a couple of leads out of bits go RG6 I had hanging around, but have ordered some RG174 leads to butcher up into the various connectors I need (TS, XLR, ETC).  We have plenty of Audio Interfaces and (audio) Isolation Transformer boxes available.

[uncle_bob]

I'm initially using this stuff to (most of all) ensure things aren't oscillating (at Higher than Audio Frequencies), and aligning the Azimuth of tape head stacks.

Have made a couple of leads out of bits go RG6 I had hanging around, but have ordered some RG174 leads to butcher up into the various connectors I need (TS, XLR, ETC).  We have plenty of Audio Interfaces and (audio) Isolation Transformer boxes available.

Hi

Keep in mind that coax comes with a "pf per foot" spec. If you have a 10 foot cable at 32 pf / ft you get a 320 pf. That capacitor loads your circuit at no extra charge. If you are probing a 1 Meg ohm circuit, it gets you a 3 db roll off at around 500 Hz.

Bob