Modifications for Crate Vintage Club 20

[martin manning]

TP1 will be inverted wrt the input. The appearance on your scope, i.e. initially rising or falling, is going to depend on how the sweep is triggered

[Redwood]

Apologies for reviving such an old thread but I'm working on one of these amps and this is hands down the most informative thread I've found about them anywhere. Before attempting any of the very interesting mods described here I want to make sure there's not something more fundamental I need to address. I've been poking around inside amps for years and successfully completed a lot of basic maintenance and repairs but only recently dipping my toe into design and modification. I've had a scope for some time but only lately getting around to really trying to implement it in everything I do. This amp is actually really motivating a lot of that just because of the great table of test point values and waveforms provided in the schematic, along with the whole Vintage Club series having a reputation for being pretty amenable to modification.

My question is right off the top though, where the waveform I pick up at the very beginning at TP1 is inverted 180 degrees from the waveform put out by my signal generator as well as that shown on the schematic. I feel stupid having to ask since I know that the waveform is supposed to flip at each gain stage, but is that true right out of the gate at the first gain stage? If so, why does the table on the schematic start with a positive slope at TP1? From reading I get the idea that cathode grounding will invert each phase but that is negated if both cathode and grid are grounded, which seems to be the case here. Obviously I'm missing something but not sure what.

I would probably just shine it on and not worry about it but frankly after replacing all of the electrolytics,screen resistors, bias resistor, and diodes all with stock values, the amp appears to function normally and is capable of being quite loud but it seems like it just has no bottom whatsoever, with the tone getting more and more unpleasantly almost ice-pick in the brain harsh the louder I turn it up. It's not so bad with the Level dimed and the Gain no higher than 1 or 2 but as soon as I start to increase the gain the treble just gets out of control and nothing I can do with the EQ makes it any more pleasing. It's not my amp so I don't really know what it sounded like before - it came to me with jacks and pots so dirty as to be almost unusable, as well as original EL84s that were just spent, all of which is fixed now. Again, no modifications at all, everything returned to stock values, double checked every installed part for correct polarity, etc. Just doesn't sound great to my ears and that inverted wave form has me wondering if there's something going on in the first stage that's got everything out of phase. Do these amps really just sound that bad in stock form?

For what it's worth, I did try a 12au7 in V1 and it made it a bit quieter but didn't seem to affect the tone that much. I also tried jumpering a 20uF cap in parallel across the first stage cathode bypass (C5) to bring it up to about 22uF and it didn't seem to make any difference at all so I didn't try anything beyond that yet. Just figured it was worth trying to better understand the scope analysis before proceeding further.

I guess one other thing worth mentioning is that I could never find an exact schematic for this particular version - the most common schematic is a later one from 1996 with 17 test points and a note that diodes D1 and D2 were deleted. The only other version I can find that includes to those diodes only has a pair of 12AX7s and no reverb. The amp I'm looking at here appears identical to the later schematic with a trio of 12ax7s and reverb, yet it still has those D1 and D2 diodes between the cathodes and plates on the power tubes. To be honest I'm not entirely sure what function those diodes are supposed to serve or what the effect was of deleting them.

[Redwood]

Thanks for confirming that. I wouldn't have questioned it but for the schematic, which has a normal positive slope at the first stage then inverts it with each successive stage. But once I got to questioning it, I read somewhere that each stage is inverted if the cathode is grounded but not if both and grid are grounded, and here that is the case, although the grid is grounded through a 1M resistor so maybe resistance that high negates that rule? Or maybe that rule is BS. I can't even find where I read it now.

Either way, this thing still sounds so harsh with ice pick highs and no bottom so I'm going to proceed to experiment with a few of the tweaks suggested above in this thread.

Thanks again.

TP1 will be inverted wrt the input. The appearance on your scope, i.e. initially rising or falling, is going to depend on how the sweep is triggered

[martin manning]

Either way, this thing still sounds so harsh with ice pick highs and no bottom so I'm going to proceed to experiment with a few of the tweaks suggested above in this thread.

These are “common cathode” stages since the cathode ground is common to the input and the output. You are correct the grid is grounded (ground referenced) through the 1Meg, but the stage is amplifying and inverting the change in grid to cathode voltage, and the 1Meg is much too large to “ground out” the signal on the grid.

In the schematic above the addition of the 68k grid resistor will pare back the highs, and the 22u cathode bypass cap will fill in the bottom end. I’d start there and see what that does.

[Redwood]

I’m still pulling my hair out over this amp. I'm not sure if proper protocol is to start a new thread in a situation like this but seems what's here already is the most comprehensive I can find regarding this amp so makes sense to go here but please correct me if that's not correct.

I’ve tried all the most conservative suggested mods:

• installed a grid stopper at V1 by replacing jumper JW1 with a 68K resistor
• added 4.7uF caps in parallel cathode resistors at R3, R6, and R12
• clipped bright cap C3
• tried increasing bias resistor R22 to first 150 then 200 ohm, no difference so returned it to 120 ohm
• Removed old solder and reflowed all the tube sockets and pots
• Very closely inspected the entire board for cold joints and haven’t found any so far, nor any damaged pads or traces
• Cleaned board with isopropyl anywhere I see any flux residue to ensure there are no shorts on the board

To recap from earlier, before even getting to the point of worrying about tone I had already done basic maintenance including:

• Replaced 30 year old C19 and C16 filter caps with new Nichicons of same value & form factor
• Replaced original B+ node filter caps with new 10uF 450Vdc along with all the other smaller cathode bypass caps with same values, all fresh quality parts from Antique Electronic Supply
• Replaced R26 470 ohm 5-watt with fresh new identical value and type
• Cleaned all pots w/ DeOxit F5 (green can) lubricating fader cleaner

I also lifted a few coupling caps to test out of circuit for leakage and found no red flags. When checking resistors for accuracy or any that might be open, I discovered R13 in the tone stack, which both versions of the schematic say is supposed to be 33.2K, was actually 68K. I assume this was from the factory because I really don’t think anybody had ever cracked this thing open before me. I happened to have a 1% rated metal film 33K so I installed that there in R13.

I’ve also tried numerous other 12ax7’s I had on hand, as well as a couple of different 12au7’s in V1. As mentioned before the EL84s are brand new JJs.

For all that effort I don’t feel like I’ve moved the needle at all. Just way too much treble, no bass at all, harsh overdriven tone at any setting other than Gain as low as I can get it without cutting off volume altogether. As bad as it is picking single notes, any attempt to strum a chord just induces a physical cringe from anybody within earshot.

I’ve gone through all 5 tubes and taken DCV measurements at each pin and don’t see anything out of the ordinary other than 21Vdc at the grid and 76Vdc at the cathode of V2b, but it’s a cathodyne PI so I don’t know exactly what DC voltages I’m supposed to see there but I know they’re not zero.

I also checked each of the schematic’s test points for DCV as well as AC peak-to-peak and waveform on my scope and while most of the DC values were maybe 5 to 10% high, most of the AC voltages were even closer than that, all except for one. The waveforms were all inverted from what was expected as described in my first post. The pure sine waves before any EQ weren’t too distorted but after that the further along I went the more distorted things began to appear – at least to my eye, but as I said I’m new at this type of analysis.

The biggest anomaly was at TP14, at the plate of V5b driving the reverb. The schematic calls for 273Vdc and I got 280Vdc. So far so good. But it wants to see 2Vp-p on the scope, and I got about 9 times that, around 18Vp-p. Note that is with all pots at 50% except the reverb, which is at 0, and the reverb cables are loaded with a 5-watt 470 ohm resistor per table setup notes.

Oh, I guess I should also mention that when measuring DC voltages on all the tube pins, I picked up an FM radio station when touching the DVM probe to the grids of V1a, V1b, and V5a (no DC voltage at any of those).

I’m still picking my way through the schematic halfway expecting to find more parts that were simply installed wrong from the factory but so far I’m only about halfway through the preamp section with that. My color coding skills suck so I have to look a lot of them up, and of course anything in parallel with other Rs or even some capacitors can throw off what I measure in circuit so I’ve actually lifted a few to verify them. I must have had this board out and back in a dozen times at this point, which creates a whole ‘nuther level of potential complications, but I am trying to be extremely careful with it. I’ve got a decent quality vacuum gun and am taking my time with any part removal and so far I don’t think I’ve damaged a single pad.

In summary, the red flags I observe are:

• FM radio coming out speaker when I put DVM probe to grids of preamp tubes
• Wave forms post tone-stack seem more distorted than they should be (but I really don’t know about that)
• Abnormally high P-P AC voltage on reverb driver tube plate but for all I know that could be a typo on schematic

What else? I've got a FLIR camera that I regularly rely on to help spot problem areas. No thermal anomalies anywhere in this circuit. Oh, and early on I tested the OT, both standard resistance tests for continuity/shorts as well as voltage testing by putting a small DC signal into the secondary and measuring the result on the primary. Nothing there that caused me to think the OT might be compromised.

I’m not a 10,000 hour guy. I have put some time in on this stuff, but generally basic maintenance and simple repairs, easy restoration stuff, not a lot of deep design work. This is the first one I can think of that seems to operate normally, almost everything appears to be in spec with a couple of exceptions I don’t really understand, but it just plain sounds bad, and nothing I can think to try seems to move the needle in either direction. I’m not even certain what part of the amp is generating the problem or how to go about injecting and monitoring a signal to isolate it.

I appreciate any or all suggestions.

[B Ingram]

... the amp appears to function normally and is capable of being quite loud but it seems like it just has no bottom whatsoever ...

... Just way too much treble, no bass at all, harsh overdriven tone at any setting other than Gain as low as I can get it without cutting off volume altogether. As bad as it is picking single notes, any attempt to strum a chord just induces a physical cringe from anybody within earshot. ...

It appears your core question is "Why is this amp so bright, and how do I make it less-bright?". We will come back to that core question, which was completely obscured by all the other items you asked & noted.

... I discovered R13 in the tone stack, which both versions of the schematic say is supposed to be 33.2K, was actually 68K. I assume this was from the factory because I really don’t think anybody had ever cracked this thing open before me. I happened to have a 1% rated metal film 33K so I installed that there in R13. ...

You can use the Tonestack Calculator to see that while there is a change in tonestack response going from 33kΩ to 68kΩ, it is not responsible for your core complaint.
In fact, the higher resistance here shifts the mid-scoop of the tonestack a bit higher, which has the effect of removing some treble compared to the 33kΩ value.

I’ve also tried numerous other 12ax7’s I had on hand, as well as a couple of different 12au7’s in V1. As mentioned before the EL84s are brand new JJs.

Tube brand & type are immaterial to your core complaint.

I’ve gone through all 5 tubes and taken DCV measurements at each pin and don’t see anything out of the ordinary other than 21Vdc at the grid and 76Vdc at the cathode of V2b, but it’s a cathodyne PI so I don’t know exactly what DC voltages I’m supposed to see there but I know they’re not zero.

There's a reason why TP9 and TP10 are where they are in the VC 20 schematic: the input impedance of a cathodyne inverter is much higher than you might guess, and very few meters can make an accurate grid-voltage measurement due to that high impedance. So the correct method is to measure at TP10, which has the same voltage as the grid (and is one reason why Crate marked TP10 on the schematic).

Your measured 76v at the cathode of V2b means it is almost certainly operating properly.

... The biggest anomaly was at TP14, at the plate of V5b driving the reverb. The schematic calls for 273Vdc and I got 280Vdc. So far so good. But it wants to see 2Vp-p on the scope, and I got about 9 times that, around 18Vp-p. ...

273Vdc = 280Vdc, near enough.

As for too much signal at the reverb driver output: probably not a big deal, and certainly not related to your core complaint.

- This might be an error on your part. All controls (Volume, Level, all EQ controls) are supposed to be at "5."
- If you have done that & you're also seeing 4vp-p at TP6, congrats, your 12AX7 might have better gain than expected.
- The 0.001µF cap (C20) feeding the reverb driver looks like 159kΩ at 1kHz (the test frequency you're supposed to use).
- C20 and R34 (1MΩ) form a voltage-divider that reduce the 4vp-p to about 3.45vp-p into the reverb driver grid.
- If you see 18vp-p out, that means your 12AX7 is delivering a whopping gain of about 5x. Since it has such a low plate load resistor, it's actually doing great.

Your tests are all supposed to have the Reverb control all the way off, so "too much drive" into the reverb shouldn't matter, and shouldn't contribute anything to your core complaint.

- If you absolutely had to "fix" this issue, it would normally be possible to break R34 into 2 resistors, and lower the signal level into the reverb driver's grid. But I suspect implementing all that would mean cutting traces in your amp, and I suspect you're not ready to take that on. Either way, the outcome would only be "less reverb."

... The waveforms were all inverted from what was expected as described in my first post. ...

You're using the waveform diagrams incorrectly. What you should be taking from the diagrams on the schematic is:
- TP2 polarity is opposite TP1's polarity
- TP6 polarity is opposite TP5's polarity
- TP9 polarity is opposite TP8's polarity
- TP10 polarity is the same as TP 8's polarity
Absolute polarity does not matter, but relative polarity is useful

- TP1 and TP2 have smooth sine waves.
- TP5 is distorting on one side.
- TP6, TP14 look jagged because of TP5's distortion, and because of the tone stack phase-shifting frequencies relative to each other.

- TP12 still has some AC Ripple (from rectification) but is is much reduced compared to TP13 because of the filtering effect of R26 and C16.


These are not "scope photographs" and you're not supposed to get too hung up on them. The chart is merely supposed to show relative sizes of DC and AC voltages at various points, and an indication of whether a waveform will be jagged/smooth, and same-/opposite-polarity with respect to some other test point.

... when measuring DC voltages on all the tube pins, I picked up an FM radio station when touching the DVM probe to the grids of V1a, V1b, and V5a (no DC voltage at any of those). ...

If you don't leave your meter-probe attached to the grid of those tubes, you will likely avoid hearing radio stations.

- What happened is normal. The meter-lead is an antenna, and you have a circuit with a lot of gain. You probably are also physically-close to some radio transmitter (whether you realize it or not).
- You probably heard AM radio, because strong RF riding on the meter-lead overdrives the triode and grid-clips one side of the signal turning the triode into a "diode detector."
- Diode Detectors are used to pick up AM radio stations, and the mechanism above is how tube guitar amps usually wind up playing a radio station through their speaker.

None of this is related to your core complaint, especially because it goes away when you're not poking a tube-grid.

... This is the first one I can think of that seems to operate normally, almost everything appears to be in spec with a couple of exceptions I don’t really understand, but it just plain sounds bad ...

Here's a wild idea: What if the Crate VC 20 just sucks?? What if you just don't like it?

______________________________________________________________

... the amp appears to function normally and is capable of being quite loud but it seems like it just has no bottom whatsoever ...

... Just way too much treble, no bass at all, harsh overdriven tone at any setting other than Gain as low as I can get it without cutting off volume altogether. As bad as it is picking single notes, any attempt to strum a chord just induces a physical cringe from anybody within earshot. ...

It appears your core question is "Why is this amp so bright, and how do I make it less-bright?".

The Crate VC 20 is a high-gain amp intended for distortion, and has a master volume in the "Level" control. Do any reading of books published about amp-design history and/or high-gain, and you'll find that designers realized that too much low-frequency content creates a mess, as harmonics of the low-notes beat against the fundamental frequency of midrange-notes. So high-gain designers generally strip lows out, and then often also shave off very high frequencies to smooth over the harsh edges of the distorted sound. High distortion also generates subharmonics that seem to fill-in the bottom end again.

- An AC15 or AC30 uses EL84 output tubes biased similarly to the VC20.
- However, an AC15 or AC30 only use a single 12AX7 triode plus a phase inverter to boost the guitar signal.
- The VC20 has 3 gain stages plus the phase inverter (and reverb, but that's not important here), so there's "way too much signal" mainly to insure the preamp will generate distortion.

So now you're encountering this amp thinking it sounds thin & harsh because "something's wrong." Except mostly, that's just the nature of the design. Because this is a single-channel amp and the one channel is optimized for the distortion-sound the designer wanted, dialing a clean sound in the amp just gives a too-bright sound. And this 1990s design pre-dates other approaches to employ switching to obtain clean & distorted sounds even when the amp doesn't contain 2 completely independent channels.

You noted the following changes:
• added 4.7uF caps in parallel cathode resistors at R3, R6, and R12
• clipped bright cap C3

C29 is another bright cap; if it remains, some low frequency content is being stripped out.

Even after you remove C29 and after adding low end with other cap-changes, you're gonna find this amp twitchy to dial-in a clean sound. And when you dial in a distorted sound, the extra low end will make that distortion batty & bad.

A final angle to try (which really should have been the first step tried) is a different speaker. I don't know anything about the speaker Crate used in these amps, but it very likely isn't great-sounding. Crate amps weren't bottom-of-the-barrel, but they were always "an affordable option" when new, and it's unlikely to have a terrific speaker. It could be you find plugging a different speaker into the amp has a positive influence on the sound.

[Redwood]

Wow B Ingram, thanks very much for this incredibly detailed reply. You've given me an awful lot to chew on, which I will try to do before coming back with further comment. Again, I really appreciate it!

[Redwood]

Well I went ahead and revoiced the preamp and phase inverter following most but not all of the changes listed in akuster777's great post from 3/6/22. I:

- halved the V1b plate load resistor at R13 to 100k
- increased the V1a cathode from 1k to 1k5
- replaced R13 in the tone stack with the original 68k I'd found there (rather than the 33.2k in the schematic)
- removed bright cap C29 (in addition to the C3 bright cap I'd already removed)
- decreased PI plate load R15 from 100k to 56k
- decreased PI cathode R18 from 100k to 56k
- reconfigured PI grid to wire C12 through new 470K with R16 at node between C12 and 470K

I'm not going to say the result is amazing and it's just a great amp now - though perhaps it really could be with a better speaker? - but it is absolutely a huge improvement such that it is at least a useful amp now, one I wouldn't be embarrassed to be heard playing. All that ice-pick-in-the-brain-harshly-out-of-tune treble has been tamed, there is now some decent bottom to it, all three EQ knobs have meaningful effect, and even the reverb sounds much better. And it's still quite loud - I didn't measure how much gain was sacrificed by the plate load resistor reductions but it ain't much.

Very sincere thanks again to all the great talent in threads like this, especially to B Ingram for so thoughtfully schooling me yesterday.

[pottedplant]

A final angle to try (which really should have been the first step tried) is a different speaker. I don't know anything about the speaker Crate used in these amps, but it very likely isn't great-sounding. Crate amps were bottom-of-the-barrel, but they were always "an affordable option" when new, and it's unlikely to have a terrific speaker. It could be you find plugging a different speaker into the amp has a positive influence on the sound.

while it is true that Crate amps are pretty cheap, I emailed Obeid Khan a long while back raving about how this amp captivated me and he actually responded. One of the things I asked him was

If you could have had your way at SLM during the production of this amplifier what are some of the things you would have included or changed? Are there corners that were cut on the production model that you would suggest be changed?

and he said

That was so long ago...

All that stuff was made the same way, actually not too bad in retrospect
Parts were decent, metal film, WIMA caps, Lee Jackson brought all those in right before I was hired,
So he actually broke them in and made them buy some good stuff.
The wave solder was the worst part. They would have inspectors that would touch up
But the pots and jacks were always suffering the cold solder joint deal later in their lifespan
PCB were pretty bad too, as cheap as they could buy.

But they did keep the Ampeg stuff up to a better standard.

So at least there is that…

I think Lee Jackson really did break them in because WIMA caps are pretty nice gear.

I might add, more specifically about the speaker itself. I did some digging and it is an Eminence model. I can't recall which one but the codes and labeling were identical to how Eminence does theirs. The model number is likely proprietary but there is a good chance it's not a custom driver but rather an off the shelf one. I can't be sure of that so take it with a grain of salt. Also I am speaking of the VC2110 which has a 10" driver in it. Wish I could say more about the 12" version but I assume they're similar. From what I can tell, Obeid is a huge EVH fan and the driver is 16 ohms in mine. Not sure if that is relevant other than the differing inductance of a 16 vs 4 or 8. I think it affects the high end?

Sorry to resurrect an older thread but I figured it would be mildly relevant. Try emailing him for anything you might want to ask, he was such a nice guy to respond he even shared a schematic with me as well as suggestions on what to change and some thoughts he had. I guess I can post the whole reply here:

Man thanks for the praise! I’m glad u dig the sound of those amps.
Bouncy is a good way to describe it – most guys like the VC3112, it is bouncy too in a different way - USA
The front end is different so the 20 watt is more biting and british based.
Yeah they come up hot and bright but that is what guitar is…mids and top end.
At least British guitar amp sound … Im a big fan
You can always play with the cathode bypass and part values to tweak it.

The bounce and reaction probably a combination of a lot of things,
Power transformer sag, EL84 tubes, cathode bias, cathodyne splitter also helps
The long tail is harder/bigger sounding.
Cathodynes will sound different for sure.
One interesting note is the 100K large-ish grid stoppers on the Output tubes
Looks like I was trying to keep the EL84’s from crossover notching when driven
Dunno if it did anything other than reducing the bandwith at the output stage

That was so long ago...
All that stuff was made the same way, actually not too bad in retrospect
Parts were decent, metal film, WIMA caps, Lee Jackson brought all those in right before I was hired,
So he actually broke them in and made them buy some good stuff.
The wave solder was the worst part. They would have inspectors that would touch up
But the pots and jacks were always suffering the cold solder joint deal later in their lifespan
PCB were pretty bad too, as cheap as they could buy.

But they did keep the Ampeg stuff up to a better standard.

So at least there is that…


Thanks again for your note man,
Hope my ramblings help you in some way,

Best,

Obeid Khan

And basically what I asked him was what contributed to the fantastic feel/reactance these amplifiers had. I surmised much much much later after learning a lot more about how amps work that it was the lack of negative feedback reducing the damping/control on the driver that I was feeling. What a feeling it has. Mine also has a good bit of awesome overtones. These are fantastic amps but are quite bright.

side note: what i would give to be able to email Dumble and ask him questions like this Email your heroes folks, never know when you won't be able to... knowing how Dumble was he probably would have called me a churl but hey maybe not....