Hiwatt Custom 20?

[absrec]

Anybody have experience working on them and/or have a reliable schematic? This is the only one I could find online, but I can tell there is at least one error in it and it has way too much gain to get a clean tone.

https://el34world.com/charts/Schematics/files/Hiwatt/Hiwatt_custom_20_tube_.gif

 

[Ó Draighnáin]

Have you seen these links and schematic?

http://el34world.com/Forum/index.php?action=dlattach;topic=8727.0;attach=13098

http://el34world.com/Forum/index.php?action=dlattach;topic=8727.0;attach=14245

 

[absrec]

Have you seen these links and schematic?

http://el34world.com/Forum/index.php?action=dlattach;topic=8727.0;attach=13098

http://el34world.com/Forum/index.php?action=dlattach;topic=8727.0;attach=14245

I haven’t. That’s interesting. No grid stopper on the input, cathode follower before the tone stack, traditional LTP phase inverter. Not too different from the typical Fender/Marshall/Vox circuit. Looks like that’s from the 70’s. I wonder if that’s the same as what is currently in production?

 

[Ó Draighnáin]

This is from a very recent Hiwatt catalogue:

"From bright spanking clean through to rich creamy overdrive - this amp can do it all! The Hiwatt Custom 20 is ‘cathode biased’ (aka Class A) which adds a ‘touch sensitive’ feel to the amps response."

I think that rules out the circuit you posted? Neither schematic shows a grid stopper. The output of the second schematic looks a lot like a Vox output. Nothing like the usual Hiwatt 12AT7 PI circuit.

I think there are quite a few Hiwatt circuits that use a cathode follower before the tone stack. The Hiwatt 50W head I built has a circuit like this. The Tone stack is also Hiwatt like.

The catalogue doesn't list an effects send in the specs for the custom 20 (does this for another model).

The input tube cathode cap for the second schematic is more in the typical Hiwatt range (nanofarad as opposed to microfarad).

 

[KLDVOX]

i think the simplest aproach would be to replace the first tube with an ecc81 this has a little less gain than the tube fitted. this saves making changes to the amp itself. if you want a lot less gain than this you could use an ecc82

 

[absrec]

The quote feature isn't working for me for some reason.

re: Cathode Bias

That's one of the confusing things I noticed. This amp is fixed bias. Upon further examination of the front panel, I noticed there is no presence control on the Hiwatt version. I built this circuit into a head that originally was JCM800/JMP style with a 6V6 power section. It has Gain-Vol-Treble-Mid-Bass-Presence so no alterations to the panel were necessary. I find the presence pretty useful since it can cut or boost (in theory anyway) so I'll keep that. The only major change I made (other than a new turret board/circuit) is to use a smaller PT. It had 660VCT (330-0-330) which, for SS rectified 6V6s, that's a little much. I have a Stancor 520VCT (260-0-260) Tweed Deluxe PT and that just puts it in a better range IMO. I couldn't find any specs for the power transformer other than one offered by TAD that's claims a secondary voltage of 250. I'm guessing they mean 500VCT (250-0-250) which is slightly lower than what I'm working with.

I did start looking at schematics for the DR103, DR504, etc. on EL34 world and noticed the trend of using a very sizable bypass cap (100-150uF) for the input preamp stage and then a 47nF bypass in the next one. I guess that's to grind the input of the 2nd stage and then trim it down. Pretty much the antithesis of a Marshall.

re: Different tube - ecc81 vs ecc83

I did try that initiallly. I put a 12AT7 in the PI and a 5751 in V1. The difference in gain was much more subtle than I expected. After staring at the schematic and looking at various other Hiwatt schematics, I decided to try cutting the second stage out entirely since none of them had that much heat on the front end. That's the way it's sitting on my bench at the moment. It sounds more 'normal' like that. I want to try to get somewhere in between. I don't mind some preamp distortion but it was a little much the way it was. I have amps that do the clean thing very well so I'm a little torn. My objective here is to put together a lower volume gigging amp. 5E3 is not enough, DC30 at 1/2 power is a bit much. This one shows potential since it's barely 13W clean on my scope. I'm wondering how much difference it would make to change the tone stack over to something more like the DR504 or if that would be a waste of time. The 100k between the slope resistor and ground is not something I've ever seen before and it's hard to say what it's purpose is. I didn't find it on any of the Hiwatt schematics I looked through.

 

[absrec]

OK… I stand corrected. It appears there were a couple models. One called the OL and the other called the Lead. Not much difference between the two. They incorporated an extra gain stage and an attenuator before the third stage marked ”overdrive”. This must’ve been what somebody was trying to create with this schematic. None of the Hiwatt schematics I found incorporated the plate load bypass cap or the Marshall style treble peaker before the extra gain stage. Anyway… Should be a fun experiment to get this thing dialed in.

 

[CJ]

$2500 price tag, sheesh,

 

[Tubetec]

Id try bypassing the first stage and see how it sounds ,
you could try adding an extra jack that interupts the signal from the V1 input when somethings plugged in , a bit like a 100w Marshall master volume,
Id go with screened cable to and from the volume pot as well , otherwise you might have unwanted interactions on the high gain settings ,
The way the amp is set up now , probably they only way you'll get clean is by backing off on the guitar volume pot .

 

[KLDVOX]

just a hunch as im no expert. but if you reduced the value of R27 increasing the global feed back wouldnt that reduce the overal gain

 

[Wilhelm]

The quote feature isn't working for me for some reason.

re: Cathode Bias

That's one of the confusing things I noticed. This amp is fixed bias. Upon further examination of the front panel, I noticed there is no presence control on the Hiwatt version. I built this circuit into a head that originally was JCM800/JMP style with a 6V6 power section. It has Gain-Vol-Treble-Mid-Bass-Presence so no alterations to the panel were necessary. I find the presence pretty useful since it can cut or boost (in theory anyway) so I'll keep that. The only major change I made (other than a new turret board/circuit) is to use a smaller PT. It had 660VCT (330-0-330) which, for SS rectified 6V6s, that's a little much. I have a Stancor 520VCT (260-0-260) Tweed Deluxe PT and that just puts it in a better range IMO. I couldn't find any specs for the power transformer other than one offered by TAD that's claims a secondary voltage of 250. I'm guessing they mean 500VCT (250-0-250) which is slightly lower than what I'm working with.

I did start looking at schematics for the DR103, DR504, etc. on EL34 world and noticed the trend of using a very sizable bypass cap (100-150uF) for the input preamp stage and then a 47nF bypass in the next one. I guess that's to grind the input of the 2nd stage and then trim it down. Pretty much the antithesis of a Marshall.

re: Different tube - ecc81 vs ecc83

I did try that initiallly. I put a 12AT7 in the PI and a 5751 in V1. The difference in gain was much more subtle than I expected. After staring at the schematic and looking at various other Hiwatt schematics, I decided to try cutting the second stage out entirely since none of them had that much heat on the front end. That's the way it's sitting on my bench at the moment. It sounds more 'normal' like that. I want to try to get somewhere in between. I don't mind some preamp distortion but it was a little much the way it was. I have amps that do the clean thing very well so I'm a little torn. My objective here is to put together a lower volume gigging amp. 5E3 is not enough, DC30 at 1/2 power is a bit much. This one shows potential since it's barely 13W clean on my scope. I'm wondering how much difference it would make to change the tone stack over to something more like the DR504 or if that would be a waste of time. The 100k between the slope resistor and ground is not something I've ever seen before and it's hard to say what it's purpose is. I didn't find it on any of the Hiwatt schematics I looked through.

For a variable-gain first stage, you can snatch some negatige-voltage from the the bias-voltage, at the output-stage-grids ( extra caps and good RC-filtering needed- as any ripple here, will become amplified and manifest as LOUD humming ) and apply it to the to first triode-cathode, via some 22-47 kohm resistor ( adjust the value for proper idle-anode-current ). Then run a front-plate-mounted 10kohm potentiometer-in-series with a 4.7-47yf cap - and connect this string in parallel to the new 22-47 kohm cathode-resistor. ( observe polarity of the cap and that it has a proper voltage-rating ) This setup would allow you to vary the gain without having to experiment with different first-stage-triodes. Best regards

 

[CJ]

try ditching V4b, go from master V into PI

12AY7/6072 for the first stage reduces gain and it is a better sounding tube,
i know, mucho $$$ ,

 

[absrec]

just a hunch as im no expert. but if you reduced the value of R27 increasing the global feed back wouldnt that reduce the overal gain

It would. This is something I've been trying to calculate as well. According to the original designs, the 50W & 100W both had a 10k NFB R from the 16ohm tap, despite the obvious differences in output voltage. It seems like a 5k6 from the 16ohm tap would give a comparable amount of NFB in this design. Comparable to the 50W that is. Unless my math is wrong.

 

[absrec]

try ditching V4b, go from master V into PI

12AY7/6072 for the first stage reduces gain and it is a better sounding tube,
i know, mucho $$$ ,

What I've read about this is -

"The magic part of the Hiwatt, and [the reason] it will produce such a powerfully clean sound, is its phase inverter. It has a bias on the grid. That bias prevents the phase inverter from shifting upward or downward, as does a typical Fender or Marshall, which gives rise to early distortion." 1 - Victor Mason, Mojave Ampworks

It's one of the main things that intrigued me about the Hiwatt design in the first place so I'm trying to keep that in the mix. I've use the straight LTP more times than I can count and it's a crowd-pleaser for sure. Just trying to do something different with this one.

 

[absrec]

For a variable-gain first stage, you can snatch some negatige-voltage from the the bias-voltage, at the output-stage-grids ( extra caps and good RC-filtering needed- as any ripple here, will become amplified and manifest as LOUD humming ) and apply it to the to first triode-cathode, via some 22-47 kohm resistor ( adjust the value for proper idle-anode-current ). Then run a front-plate-mounted 10kohm potentiometer-in-series with a 4.7-47yf cap - and connect this string in parallel to the new 22-47 kohm cathode-resistor. ( observe polarity of the cap and that it has a proper voltage-rating ) This setup would allow you to vary the gain without having to experiment with different first-stage-triodes. Best regards

I will have to read up on that. Until I understand what I'm doing, I'm a little apprehensive to do it. I feel like I've seen something similar to what you're describing. Maybe even on one of the Hiwatt schematics I looked at (?). They tried so many different things.

 

[absrec]

DR504OL

I ended up doing something more along these lines. Only with a 6V6 output section. Interestingly, it not that different from the first schematic I posted. Still 4 gain stages before the PI, just different value cathode bypass caps. I removed the plate load bypass on stage 2 as well as the 470p "treble peaker" directly following it. The main things I'm curious about is R7 on the first schematic I posted. It's 330k. I changed it to 47k like the one linked to here. It's basically in parallel with the attenuator. It wouldn't serve as a grid leak, AFAIK. I feel like changing that value had a dramatic effect on the overall gain. Meaning, it came down to a reasonable level. It's sounding pretty good now. I know it can be better though.

I'm gathering up ideas and just got some parts in the mail I was waiting on so I plan to dive back in over the next couple days. I'll report back.

 

[Wilhelm]

It would. This is something I've been trying to calculate as well. According to the original designs, the 50W & 100W both had a 10k NFB R from the 16ohm tap, despite the obvious differences in output voltage. It seems like a 5k6 from the 16ohm tap would give a comparable amount of NFB in this design. Comparable to the 50W that is. Unless my math is wrong.

Tube-amps don't work like ic op-amps. Don't expect normal function from it - if you do drastic changes to the NFB-loop. Remember there's a chunky transformer at the output - which is guaranteed to introduce phase-shifts outside it's normal bandwidth - . Doubling the NFB is likely to push the little guitar-combo into oscillation-mode, with all kinds of interesting sonic results

 

[KLDVOX]

altering the feedback can be a good move. marshall vertually copied the fender bass man. but took the feedback from a different tap on the output transformer and got a different amp itsals a very easily reversed mod.

 

[CJ]

There is very little phase shift on a power output transformer as there is plenty of current to charge the parasitic capacitance, most Schumacher's are good past 100 K.
I think the reason I got so much phase shift on that Model T over at the lab is because I did not disconnect the red lead from the power supply.

Fender Deluxe has no NFB.

5150 has 5 gain stages but there are voltage dividers all over the place.

 

[absrec]

Yeah.... NFB ain't rocket surgery. A good mod on Fender amps is to remove it entirely. Rewire the polarity switch (after removing the death cap and fitting a 3 prong, of course) so it switches the NFB resistor in and out. It helps it to sound more exciting at lower volumes. To be honest, I usually prefer less NFB anyway. Too much and the amp starts to sound vanilla with it's even frequency response and clean headroom. Boring.