Measuring the impedance at the FX loop

[Stephen1966]

Greetings,

I am trying to determine the impedance at my FX loop. And a couple of tests this evening produced some interesting results. My loop is passive, and modelled after the #124. Mine, isn't exactly the same, but the testing and measurement are probably good for all the ODSs with passive loops.

First, I'm using a scope to find Vrms with the probe, GND (-) and signal (+) attached as shown in the schematic:

#124 FX loop.jpg

This is just the first stage, of course - to find the unloaded Voltage (Vrms) before I add a load and then do the sums.

My readings:

when the scope is DC coupled are, 2.48 Vrms / 4.98 Vpp
when the scope is AC coupled are, 19.3 Vrms / 59.6 Vpp

I would like to confirm, first of all, if it is the AC or the DC we are interested in.

This photo tells us MrD measured 2.5 Vrms

ODS_124091.jpg

Am I correct in assuming this is with a scope DC coupled?

I'm using a sine wave at 1kHz/ ~1V as the input signal at the front of the preamp and these Vrms and Vpp readings were taken with the Volume and Gain maxed out. Briefly, because my dummy load got as hot as hell!

If anyone could clear away the mystique of how MrD arrived at his figures, it would be much appreciated.

Stephen

[pdf64]

…
I'm using a sine wave at 1kHz/ ~1V as the input signal at the front of the preamp and these Vrms and Vpp readings were taken with the Volume and Gain maxed out…

What’s the rationale for overdriving it so hard?
I suggest to adjust stuff to keep it in its linear range, worst case. So master set electrically half way.

[Stephen1966]

…
I'm using a sine wave at 1kHz/ ~1V as the input signal at the front of the preamp and these Vrms and Vpp readings were taken with the Volume and Gain maxed out…

What’s the rationale for overdriving it so hard?
I suggest to adjust stuff to keep it in its linear range, worst case. So master set electrically half way.

First, just to see what the range is. A quick test, that's all. Granted, the operating level isn't even half that. It comes to life and wakes the dead with Master and volume at around 40%. It is maybe something of a coincidence that it maxes out at 2.48 Vrms when the scope is DC coupled. What is the thinking with MrDs figures then? If they indicate an optimal operating voltage (AC coupled) then we are a step closer to understanding the signal settings - volume and master settings - which might then be used with the adjustment of the OD and PI trimmers and a possible host of other tone shaping adjustments. I think most pedal effects will have a blocking cap and relatively high impedance (1Meg) on their input so it's the AC that gets through. And 60 Vpp would fry most pedals, I'm sure. If it wasn't some operational sweet spot, then could it be a reference to dBu? 2.5 Vrms is just about 10 dBu. I mean, what do you think?

[ijedouglas]

Stephen,

I think you are correct with your assumption on max controls. I would think that the 2.5Vrms would indicate the max output of the effects send.

[pdf64]

I don’t get why the DC and AC coupled readings are different. Surely there’s no DC feed into the master vol?
Could you upload scope screenshots?
Maybe an asymmetric waveform is causing an effective DC offset?
Need more schematic and operating point details of the amp really.

[martin manning]

My readings:

when the scope is DC coupled are, 2.48 Vrms / 4.98 Vpp
when the scope is AC coupled are, 19.3 Vrms / 59.6 Vpp

2.48 Vrms is 7.9 Vp-p, and 19.3 Vrms is 54.6 Vp-p for an undistorted sine wave. You must have some clipping going on.

I would like to confirm, first of all, if it is the AC or the DC we are interested in.
Am I correct in assuming this is with a scope DC coupled?

It's AC you are interested in (the output is quoted in Vrms), but the scope coupling shouldn't matter. You will see the same AC either way. I'd guess the markings are indicating that the output level is 2.5 Vrms, and the expected load impedance is 500k, and this is probably assuming a low-distortion sine wave.

[rootz]

What probes are you using? There seems to be something loading your signal. Might be a 1x probe with low impedance?

Anyway, with 1V p-p input, volume and master at max, no PAB, clean output alone should be more like 200V p-p (70V RMS) IIRC. Pretty much full possible swing around a 200V plate voltage and 300V B+.

Please correct me if I'm wrong about this.

[Stephen1966]

My readings:

when the scope is DC coupled are, 2.48 Vrms / 4.98 Vpp
when the scope is AC coupled are, 19.3 Vrms / 59.6 Vpp

2.48 Vrms is 7.9 Vp-p, and 19.3 Vrms is 54.6 Vp-p for an undistorted sine wave. You must have some clipping going on.

Hi Martin, can I just check... are you using Vpp = Vrms (2√2)? In any case, the calculations don't tally with the measured Vpp and it's fair to say there is clipping. Somewhere in the first stage of the preamp, at or prior to the volume pot, I suspect, since adjusting the gain increases the signal without clipping.

Here are the traces for Volume and Master at 100% (the blue trace is the input signal at the front of the amp, the yellow trace - CH1 -is the measured signal) :

Full DC probe (x1).jpg

It's AC you are interested in (the output is quoted in Vrms), but the scope coupling shouldn't matter. You will see the same AC either way.

Now, it gets interesting, because I couldn't repeat the DC reading I got last night (maybe I was reading it separately from the DMM!?) Anyway, staying out of the DC rabbit hole for now, here is the same setting when the channel is AC coupled:

Full AC probe (x1).jpg

I'd guess the markings are indicating that the output level is 2.5 Vrms, and the expected load impedance is 500k, and this is probably assuming a low-distortion sine wave.

That makes sense to me. I mean, it makes sense that you would want a nice clean signal going into any pedal effect so I played around with volume and master settings, turning them down but keeping them more or less at the same relative position on the dial, but out of curiosity here is what my signal looks like at ~2.5 Vrms:

2.5Vrms.jpg

I am measuring 5.48 Vpp from 2.47 Vrms - the theoretical Vpp should be closer to 7 Vpp.

Effects with higher impedance, which I suspect are more common these days, suggest the same signal could be at a lower Vrms. It doesn't want to be too low though, or it starts to pick up noise from the sound floor.

Here then, is a nice clean signal from what I consider to be a quiet playing level. Not too quiet, about practice level:

Low volume.jpg

And for Vrms of 818.17 mV the measured 2.34 Vpp is slightly higher this time, than the calculated 2.31 Vpp.


Dialling up the Volume and the Master to what I consider a normal playing level, about 40% - it's pretty loud - this is what I get:

Normal volume.jpg

This is just after the onset of clipping and for a 1.34 Vrms I measure 3.08 Vpp when it should be around 3.7 Vpp.

So, it's probably fair to say, that the passive loop shouldn't be driven much harder than 2.5 Vrms which isn't going to exceed around 7 Vpp - safe territory for the majority of 9V powered effects. The level of the volume affects the amount of clipping so alongside the master to keep amplitude of the signal at an operational level, it offers up a lot of tonal shaping.

[Stephen1966]

I don’t get why the DC and AC coupled readings are different. Surely there’s no DC feed into the master vol?
Could you upload scope screenshots?
Maybe an asymmetric waveform is causing an effective DC offset?
Need more schematic and operating point details of the amp really.

Hi Peter,

Theoretically, I guess the DC offset I encountered was an attribute of pulsed DC coming off the plate of the second stage, lifting the AC. Anyway, in practice, I haven't been able to reproduce the DC measurement, Martin was right, I saw the same AC voltages either way and as that is what we are interested in, its an anomaly that there appeared to be some DC leakage if that is what it was. I'm pretty sure my coupling cap between the second stage and the master is okay.

My schematic is a work in progress and the FX loop circuit is a prototype that has had some problems I've been working through. I could post the whole thing here, but if you could tell me what parts of the schematic and more about the operating point details you need, I could post those in simpler, more focused form.

From the scope traces I've posted, it's clear to see the waveform is indeed asymmetric although it only becomes noticeable later on.

[Stephen1966]

What probes are you using? There seems to be something loading your signal. Might be a 1x probe with low impedance?

Anyway, with 1V p-p input, volume and master at max, no PAB, clean output alone should be more like 200V p-p (70V RMS) IIRC. Pretty much full possible swing around a 200V plate voltage and 300V B+.

Please correct me if I'm wrong about this.

Hi rootz,

Up to reading your post - probes at 1x. The impedance built into the scope, is 1M as you can see in the scope screenshots. Your settings are pretty much what I used as well. Around 1 Vpp input, no PAB (or OD), switches in clean positions, tone controls and presence at midnight on the dial. I think if I were to put the probe on the plate of the second stage I would get somewhere in the ballpark of your voltages but I'm interested in what the FX loop would see so I've been taking my readings from the wiper of the master.

Between the second stage and the master there it the tail end of my reverb circuit. Specifically, from the plate, the signal passes through the .05 coupling cap, clean through the OD relay then along to the reverb circuit (dry side) where it meets a 470k mixing resistor and another 470k resistor which is part of the LNFB on the last stage of the reverb circuit, and then (finally) on to the j-taper, 1Meg MASTER pot. The tail end of the dry mixer side of the Rvb also has a 470k grid leak resistor to ground. Nothing too controversial about that, you might say, but if I wanted loading, I'm pretty sure I got it

I didn't think there was anything too problematic with the 1X attenuation of the probe. Interestingly, with the probe at 10x attenuation, here is what I got:

Full AC probe (x10).jpg

And you can see now my 60 Vpp has dropped to 53.6 Vpp with the same max settings on the amp as before. The 1x reading above, and the 10x reading posted here, were taken within seconds of each other, so it isn't likely there was a significant change in the mains voltage though I do see some big swings in my home. Anyway, it appears the Vpp has gone the opposite way this time: for 18.04 Vrms, I calculate I should have got 51.02 Vpp. Instead, I got 53.6 Vpp.

[sluckey]

If you set your input level to a realistic level of about 100mVpp your waveforms will look nicer, and the math may work better.

[Stephen1966]

If you set your input level to a realistic level of about 100mVpp your waveforms will look nicer, and the math may work better.

Thanks, Steve, do you mean the input level of audio generator at the front of the amp? 100mVpp is pretty low, even for guitar pickups.

[Stephen1966]

I suspect all these words I'm using aren't making things any easier to understand

Here's the amp in schematic form as it stands at the moment:

SKYLINER 2.5.pdf

[sluckey]

If you set your input level to a realistic level of about 100mVpp your waveforms will look nicer, and the math may work better.

Thanks, Steve, do you mean the input level of audio generator at the front of the amp? 100mVpp is pretty low, even for guitar pickups.

Yes. 100mVpp is a good input level for a guitar amp. 1Vrms is line level, not instrument level. 1Vrms is what you would want to send to a power amp. It's way too hot to send to the instrument input. Look at what Fender typically uses for sig gen input to their reissue AB763 amps...

https://el34world.com/charts/Schematics ... manual.pdf
https://el34world.com/charts/Schematics ... sue_sm.pdf

[Stephen1966]

If you set your input level to a realistic level of about 100mVpp your waveforms will look nicer, and the math may work better.

Thanks, Steve, do you mean the input level of audio generator at the front of the amp? 100mVpp is pretty low, even for guitar pickups.

Yes. 100mVpp is a good input level for a guitar amp. 1Vrms is line level, not instrument level. 1Vrms is what you would want to send to a power amp. It's way too hot to send to the instrument input. Look at what Fender typically uses for sig gen input to their reissue AB763 amps...

https://el34world.com/charts/Schematics ... manual.pdf
https://el34world.com/charts/Schematics ... sue_sm.pdf

Thank you, I knew you are wise

And yeah! [Second fender service manual] There it is! '37mVAC' at the grid of the first stage. Unless I am missing something though (always possible) I understand this to mean the RMS value, not peak-peak. Have to say that even so, this doesn't reflect my pickups https://ampgarage.com/forum/viewtopic.p ... 46#p440446 There, I tested my pickups in both parallel and series with some fairly aggressive (but still not absolutely mental) strumming and with my pickups in parallel I averaged out at around 330mV(rms) (3.36Vpp measured with some harmonics - I know the maths don't agree!), but in series, closer to 800mV (6.24Vpp measured - again, its literally peak to peak, no matter how transient those peaks were).

I'm giving it a shot at 100mVrms though. I (sort of) get what you are saying how it might clean up the signal... and the maths. The lower voltage, coupled with the low(ish) frequency of 1kHz reduces the THD in my understanding. Okay, it doesn't match my pickups, but so what It is as much about getting as clean a signal as possible even if the input is only a fraction of the level we will be using. A baseline, doesn't have to be the playing level, right?

[Edit: apologies - a momentary lapse of stupid! - 37mVAC (rms) is around 100mVpp!]