Crossover distortion scope shots and questions

[waspclothes]

Hey guys,
I have an amp that seems to have a rather "harsh" distortion at higher volumes, so I hooked up an oscilloscope to try and see what's going on.

The amp is a silverface deluxe reverb, and I'm using a 1Vpp 2khz sine wave going into the 1 input of the vibrato channel. The speaker output is going to an 8 ohm dummy load and the other channel of the scope is connected to the speaker output.

Here's a picture of the 'scope with the volume on the amp cranked to 10.



Is this the typical "crossover" distortion I should be seeing? FWIW, the signal looks pretty pure until the volume knob gets to about 5 or so, and then clipping occurs at the top of the wave as well as the subtle crossover points appearing and then becoming very spikey as the volume is increased.

[Audiotexan]

I'd be looking into the bias section as the very first step.

Also, is this the version with dual 6V6's and a 5U4 with a volume boost?? If yes, then the bias starting point should be approximately -35vdc. (and tailor to the amp/tubes from there)

[Audiotexan]

Normally, crossover distortion notches should look like the following (and that's still a bit much IMO):

Picture credit to user SwedishWings at diyaudio.com

[waspclothes]

I'd be looking into the bias section as the very first step.

Also, is this the version with dual 6V6's and a 5U4 with a volume boost?? If yes, then the bias starting point should be approximately -35vdc. (and tailor to the amp/tubes from there)

Yes Audiotexan, this is the 5U4 with the volume boost. The bias output is -51VDC with the tubes biased at 460V plate voltage and 17mA idle current - for about 60% dissipation.

[waspclothes]

Normally, crossover distortion notches should look like the following (and that's still a bit much IMO):
[ATTACH=CONFIG]29031[/ATTACH]
Picture credit to user SwedishWings at diyaudio.com

Jeez, yeah.. that looks a lot cleaner than my scope shot. With the volume below 6 though, my scope shot does look very clean, those crossover points only start showing after 6 and then start to get very spikey after that. I also notice that the crossover distortion on my scope shots aren't at all symmetrical, unlike the screen you posted.

[eschertron]

Yes Audiotexan, this is the 5U4 with the volume boost. The bias output is -51VDC with the tubes biased at 360V plate voltage and 17mA idle current - for about 60% dissipation.

Are you measuring 17ma PER TUBE or for both tubes? -51vdc is pretty c-c-c-cold

[waspclothes]

Are you measuring 17ma PER TUBE or for both tubes? -51vdc is pretty c-c-c-cold

17mA per tube.

[Audiotexan]

Yes Audiotexan, this is the 5U4 with the volume boost. The bias output is -51VDC with the tubes biased at 360V plate voltage and 17mA idle current - for about 60% dissipation.

Well, judging by the formula (360 x .017 = 6.12w) something doesn't seem quite right. If it's in triode mode, then yeah, that'd be 60%, but if it's in pentode mode, then that's less than half of rated Dp of 14w(at least using the JJ Spec sheet for a 6v6). -It's not the first thought for most, that someone might have changed the wiring/mode on the output tubes.

I'd say closer to 19-20ma should warm things up a bit, and put it closer to the 70% marker. So to me it still sounds a bit cold.

If you go here and view the BiasChart-6V6.jpg..You're kind of on the high end of the spectrum voltage-wise according to that chart (even though the schem shows it to be 415vdc sourced).

So this brings up a few points to check.
One, the rectifier tube is getting weak. (as you've got 360 out of 415)
Two, the bias is still a touch off.

[Audiotexan]

Yes Audiotexan, this is the 5U4 with the volume boost. The bias output is -51VDC with the tubes biased at 460V plate voltage and 17mA idle current - for about 60% dissipation. *Last edited by waspclothes; Today at 08:49 PM.*

Ok..noting your change to the B+...that at least removes the potential rectifier issue I mentioned. Good news!

However, you didn't change your stat on plate dissipation (60%). So I'm guessing by your edit that would mean that you're now running at 7.82w yes?

Until you confirm that it's still wired 'factory'...I'll leave that aspect alone, and assume that you're just running really cold.

Just in case you don't have the schem for that one handy...SF_Deluxe_Reverb_Boost.zip

Also, if you edit something that's "critical" to the process, it's always nice to note such, that so that someone coming in to the thread after the fact has a heads up As I had to do a double take and wondered how I'd missed that 100v difference until I verified it by eschertron's quoting you. lol

Cheers!

[waspclothes]

Well, judging by the formula (360 x .017 = 6.12w) something doesn't seem quite right. If it's in triode mode, then yeah, that'd be 60%, but if it's in pentode mode, then that's less than half of rated Dp of 14w(at least using the JJ Spec sheet for a 6v6). -It's not the first thought for most, that someone might have changed the wiring/mode on the output tubes.

I'd say closer to 19-20ma should warm things up a bit, and put it closer to the 70% marker. So to me it still sounds a bit cold.

If you go here and view the BiasChart-6V6.jpg..You're kind of on the high end of the spectrum voltage-wise according to that chart (even though the schem shows it to be 415vdc sourced).

So this brings up a few points to check.
One, the rectifier tube is getting weak. (as you've got 360 out of 415)
Two, the bias is still a touch off.

Sorry, a few minutes after I posted this I realized I typed 360V instead of 460V for the plate voltage.

Just out of interest, I tried a different pair of power tubes biased to the same % and got the same crossover distortion when the volume was put up past 6.

[JoeM]

-51v bias is more in 6L6 territory. How are you measuring the bias current?

[Audiotexan]

Sorry, a few minutes after I posted this I realized I typed 360V instead of 460V for the plate voltage.

No harm done!
Like I said, I just a had a 'head scratcher' moment and was like 'Did I really miss that?? ...where the hell did I come up with 360 then?' lol

Just out of interest, I tried a different pair of power tubes biased to the same % and got the same crossover distortion when the volume was put up past 6.

Same brand/batch of tubes should yield the same results. All else remaining the same. (I think the only way that would change enough worth noting would be going from using NOS tubes to modern, or vice-versa)

I'm still of the opinion that you'll see that notch disappear if you get nearer the -35 to -40 range, and set between 20mv-24mv =)

Edit: assuming pentode operation of course!

[52 Bill]

The amp is a silverface deluxe reverb, and I'm using a 1Vpp 2khz sine wave going into the 1 input of the vibrato channel. The speaker output is going to an 8 ohm dummy load and the other channel of the scope is connected to the speaker output.

1 volt peak to peak input?

[g1]

As far as looking at the crossover notch, there is no point looking after the signal is clipped. You can take the notch right out and it may come back at (or beyond) clipping. It will be different for various circuits.
Once you are hard clipping like that, you probably won't discern which part of the distortion is the crossover distortion.
Do you have a shot of what the notch looked like before clipping?

[waspclothes]

No harm done!
Like I said, I just a had a 'head scratcher' moment and was like 'Did I really miss that?? ...where the hell did I come up with 360 then?' lol



Same brand/batch of tubes should yield the same results. All else remaining the same. (I think the only way that would change enough worth noting would be going from using NOS tubes to modern, or vice-versa)

I'm still of the opinion that you'll see that notch disappear if you get nearer the -35 to -40 range, and set between 20mv-24mv =)

Edit: assuming pentode operation of course!

As I had the scope hooked up I actually changed the bias current to as high as 35mV and watched the distortion change. It did make a difference, and I was able to maybe get the volume up to 6.5 (instead of 6) before it started noticeably distorting at the crossover points. But once at 10, it was pratically the same.

I've never hooked up a scope to another amp and fed it a 1Vpp sine wave before, so I don't even know if this is normal or not - I guess that's a big part of my question. What I do notice is that the crossover points only start to distort once the volume gets high enough that the waveform starts clipping at the top/bottom as well, so both distortions are happening at the same volume level anyways. That's why I'm sort of wondering now if is a "problem" at all, since either way I'd be hearing distortion at that point. I'm just not sure if the spikeyness of the crossover point distortion would be considered to sound less pleasing (and explain the "buzzy" distortion I seem to be hearing and not liking), or if it's just what the amp sounds like when it distorts.

I also scoped the screen of each 6V6 and they're receiving nice looking sine waves, even with the volume on 10.

I'll check to see if the amp is in fact stock, I didn't see any changes/bodges/etc so just assumed.

52 Bill> Yes, 1Vpp sine wave @ 2khz is what I'm feeding into the amp. Let me know if that's too high or low, I just attached the scope to a guitar and strummed a few chords to see the voltage output of that and it seemed to be around .8-1.2Vpp depending on how hard it was strummed.