Modifications on Hoffman Plexi 50, sanity check

[uki]

Hey guys!!

I have built this amp back in 2017, and although it works, it does go on overdrive berserk too easily too soon.

So after sometime searching and thinking and overthinking about it, here what i came up, to move the tone stack between the V2 triodes,

I got a few questions:

 - The new position of the tone stack is properly set, and if R15 value on V2b is ok?!
 
 - Bias adjust/balance, will those resistor and pot values work on this amp?(circuit from Rob Robinette)

Thank you in advance!


PS. the thread went to a different path.

[stratomaster]

Before completely changing the topology of a proven design to something untested I'd encourage you to try something like a 220k resistor on the input lugs of the volume pots and/or a 2.2M-4.7M resistor to ground at the grid to V2A. Even a pre phase inverter master volume after the treble pot can drastically reduce the breakup.

[tubeswell]

So methinks getting a plexi sound with the same number of tubes is a matter of having a direct-coupled pair (inverting gain stage and DC-coupled CF) driving the tone stack (and omitting the 'tone stack recovery stage between the TS and the LTP), which is essentially a JTM45 approach. And then if you still want more fizz, taking the V1 tube and cascading that instead.

[tubeswell]

E.g. as in Ceriatone Yeti. (Son of yeti schematic - i.e. with 6V6s - shown)

[uki]

Thanks for the reply guys!

@stratomaster  I will do the suggested modifications and report back

@tubeswell  i dont mind modifying the amp to something else as long the amp get clean tone fender like or similar,
not so fan of marshall( i tried not my thing)

That idea to move the tone stack i got from the marshal major, and mostly from fender amps,
I want the amp to have a clean sound at high volume, lots of headroom

also i found this local feedback loop from this video(see picture), what do you guys think?

[tubeswell]

dont mind modifying the amp to something else as long the amp get clean tone fender like or similar,

Remove the tone stack recovery triode and use that for a cathode follower connected to the triode before the tonestack. That will decrease the preamp gain and sound more like a 1959 Fender tweed Bassman.

[uki]

 

>which one is the tone stack recovery triode ?
Is it in the schem in the first post? that is a modified schematic, not the original from here which is the way the amp is.

[tubeswell]

>which one is the tone stack recovery triode ?
Is it in the schem in the first post? that is a modified schematic, not the original from here which is the way the amp is.

triode after the tone stack (but before the LTP) - i.e., it's 'recovering' gain that has been lost in the TS. In this case it's not needed - especially if you want a cleaner signal.

[uki]

triode after the tone stack (but before the LTP)

That is only in the modified schem, the amp still as it is in the original schem, i did not move anything around yet.

But would moving the tone stack as in the modified schem would reduce gain? Also crossed my mind making it a twin like preamp.

I did not yet had time yet to do @stratomaster suggestions

[tubeswell]

triode after the tone stack (but before the LTP)

That is only in the modified schem, the amp still as it is in the original schem, i did not move anything around yet.

But would moving the tone stack as in the modified schem would reduce gain? Also crossed my mind making it a twin like preamp.

I did not yet had time yet to do @stratomaster suggestions

Can you link to the schematic you're woking from (to save everyone else having to spend time hunting for it)?

[uki]

Can you link to the schematic you're woking from ?

I did 

the original from here which is the way the amp is.

the schem in the 1st post is just an idea, if it is possible to do it

[tubeswell]

the original from here which is the way the amp is.

So that's what I'm talking about w.r.t. something more like a Marshall JTM45/Fender 5F6A. It should give you a cleanish sound until you dime the volume past 6 or 7. If it doesn't, then something is wrong. Or if it is 'clean' and you want it 'cleaner' - try swapping a 12AT7 into the LTP slot

[uki]

It should give you a cleanish sound until you dime the volume past 6 or 7.

That would be great atm amp goes on berserk OD at 3 or so

If it doesn't, then something is wrong.

We did a trouble shooting while back (here) and no clue about what it could be,
methinks its too much gain all over the circuit, specially V2 circuit(cathode follower), if it go good , that is my though on modded schem! 

Or if it is 'clean' and you want it 'cleaner' - try swapping a 12AT7 into the LTP slot

I tried , no effect.
The amp is far way from clean...

[tubeswell]

Is each filter cap node decoupling the power supply properly? (Separate HT supply nodes should be at different voltages)

[uki]

I got some numbers  :

bias -55v

Nodes A 478v , B 478v , C 333v , D 285v , E 267v

I tried @stratomaster suggestion, the resistor from V2a grid to ground, 2.2M and 3.9M the ones i have. can notice a little difference with 3.9 but far from clean.

Here the question again, moving the tone stack to between V2 triodes and eliminating the cathode follower, would it reduce the harsh gain?

[tubeswell]

I got some numbers  :bias -55vNodes A 478v , B 478v , C 333v , D 285v , E 267v

Those look normalish, except, if that's the bias voltage at the output tube control grid pins, it may be biased way too cold.


Which could cause crappy distortion - what is the cathode current looking like in each output tube?

Here the question again, moving the tone stack to between V2 triodes and eliminating the cathode follower, would it reduce the harsh gain?

Not likely. Look at the (potentially cold?) bias issue first

[uki]

Moving the tone stack at 1st glance looks easy enough to implement without changing much the circuit. Still just an idea,
the second idea is to go twin like fender.

the bias voltage at the output tube control grid pins, it may be biased way too cold.

that is the maximum negative  -55   

Not likely. Look at the (potentially cold?) bias issue first

We have done it before, bias at like -42 , amp still have lotta distortion and one of the PA tubes did red plate when going a bit above that.


BTW you asked me years back if i had an scope, well now i have, not sure how to use it thou 

[tubeswell]

what is the cathode current looking like in each output tube?

[uki]

what is the cathode current looking like in each output tube?

Meter at 200mV DC 
 
V4 1.5mV

V5 2.4mV

bias at -55.

[stratomaster]

I tried @stratomaster suggestion, the resistor from V2a grid to ground, 2.2M and 3.9M the ones i have. can notice a little difference with 3.9 but far from clean.

That forms a voltage divider. You would have heard a cleaner sound with the 2.2M, not the 3.9M. The 2.2M passes less signal to the following grid than the 3.9M.

Your amp is unhealthy. Before you modify it you need to identify the fault.

Edit: reading on in the thread-- your bias is way too cold.  You're getting crossover distortion.

[tubeswell]

what is the cathode current looking like in each output tube?

Meter at 200mV DC 
 
V4 1.5mV

V5 2.4mV

bias at -55.

Way too cold


you want a good 30mA in each tube - and if an output tube is red-plating at 30mA, you've got a bad tube

[uki]

Your amp is unhealthy. Before you modify it you need to identify the fault.

We been through it and no conclusion on that, im willing to give it another go.

Edit: reading on in the thread-- your bias is way too cold.  You're getting crossover distortion.

you want a good 30mA in each tube - and if an output tube is red-plating at 30mA, you've got a bad tube

Even with a not so cold bias the amp does distort too much too soon.

One of the reasons i think a bias balance would help. Tubes are not very well match seen so.

Gonna change bias and report back.

[uki]

Ok went up with bias

Now V4 at 27mV and V5 30mV

No red plate. (i think back when it did red plate i went a bit too much on bias, if i remember right)

The tone got better but only until the amp start distorting at like volume 3, beyond that its just more distortion.

[tubeswell]

Can you post a soundbyte sample of the distortion?

It must be a wiring mistake, bad component or bad tube.

Maybe try tube rolling in the preamp first.

If that doesn’t work, scope the signal chain from front to back with a 1Khz signal at the input (to find where it’s happening)

[uki]

Can you post a soundbyte sample of the distortion?

Yeah I'll do that first

Maybe try tube rolling in the preamp first.

that will be the second take

It must be a wiring mistake,

I did check all the connections, the amp works, all controls works,
Now about the wiring it is really old wires cloth type used in the amp, been thinking about replace all the old wires.

bad component or bad tube.

Could be a bad component, some are really old, caps and resistors, we did check the critical caps while back no leak on those,
if it is a tube the swaping may spot something

If that doesn’t work, scope the signal chain from front to back with a 1Khz signal at the input (to find where it’s happening)

How(ground/signal path?) and at what points in the amp to connect the scope(its a really cheap one/probe x1 x10)?

[tubeswell]

Don't worry about the scope until you've tried all the other stuff

[uki]

Ok here is the sound sample. also in the begin of the video the preamp tubes are visible, both sets of tubes give same behavior with tubes swap. In the video i tried to set both volume controls at about 2, 3 and 4, 1 is clean nice and clear tone, 2ish have a little saturation,  beyond it distortion start.

[stratomaster]

Are you using audio taper volume pots? Also, I only saw 1 input. Do you have a schematic of how your did the input stages? If you cascaded them (like the hot switch option) then it's performing as intended.

[uki]

The input is just like in the schematic in the first post here.

Are you using audio taper volume pots?

Im not sure what type they are. those two are really old vintage,

here some images

[stratomaster]

Taper absolutely matters on volume pots and correlating saturation and volume settings.

Most use a 10% audio taper.  This means that what is normally halfway up on an audio pot is equivalent to just a few degrees of rotation on a linear one.

I recommend you disconnect wires to the pot and measure the resistance from the center lug to each outer lug with the pot set mid way.

[uki]

Here is the ohms values in the pots:

[stratomaster]

Here is the ohms values in the pots:

You have the wrong pot taper. Linear pots will have next to no adjustment in the clean range. Sub in audio taper and I think you'll get the clean range you're looking for. If not, we can keep troubleshooting

[uki]

I got a pair of Alpha pots type A , ill give a try on that and report back.

[uki]

Pots replaced, there is good improvement the harsh distortion isn't happening as soon as before. 1 to 4 clean ish, beyond it start to drive.
The amp still very agressive, very grit. I have tried again the mod suggest by stratomaster, 2.2M resistor in the V2 pin 2 to ground, it did help a little more. Amp still need a lot more clean tone and headroom.

Im willing to modify it to achieve the clean clean tone.

[stratomaster]

Pots replaced, there is good improvement the harsh distortion isn't happening as soon as before. 1 to 4 clean ish, beyond it start to drive.
The amp still very agressive, very grit. I have tried again the mod suggest by stratomaster, 2.2M resistor in the V2 pin 2 to ground, it did help a little more. Amp still need a lot more clean tone and headroom.

Im willing to modify it to achieve the clean clean tone.

remove the 0.68μF cap from the 2nd stage. You'll probably want to dial in less of the normal channel and a touch more of the high treble channel after doing this to keep a comparable eq.

[uki]

remove the 0.68μF cap from the 2nd stage.

Would that be C1b(V1) or C4b(V2) ?

[stratomaster]

C1b is the first stage. You just put them in parallel.

[uki]

C4b is now out of the circuit, the bright channel now is much more clear and only start drive a bit after 5. could have a little more punch.

The 2.2m resistor is still in place.

The normal channel is too low frequency, bassy and dirty, same as before.

[stratomaster]

Cool. Now we're getting somewhere.

Take the 2.2M out and swap C2 and C3. Just remove both and exchange their places.  Put the 0.0022 on C2 and the 0.022 on C3.

[uki]

With caps swapped, bright channel got punchier(i like it) but lost clarity, no so clean anymore, i guess the ideal value somewhere between the 2 caps but going to the side of .022

Normal is still way too bassy. Maybe it is that cathode cap 330 too much?

Tested with and without 2.2m resistor, with still better, without gets dirty grit tone.

[stratomaster]

With caps swapped, bright channel got punchier(i like it) but lost clarity, no so clean anymore, i guess the ideal value somewhere between the 2 caps but going to the side of .022

Normal is still way too bassy. Maybe it is that cathode cap 330 too much?

Tested with and without 2.2m resistor, with still better, without gets dirty grit tone.

The complaints you have about this amp are what makes people so attracted to it. The early breakup, gritty and bright lead channel and bassy normal channel are what made this circuit legendary.  The way I like to use this circuit is jumping the two, turning up the lead channel, and just barely feathering in the normal channel to help fill out the low end--essentially using the normal channel volume as a bass control.

I think you'll benefit from a pre phase inverter master volume--which is just a 1M to ground off the treble control with the wiper feeding the input cap to the phase inverter. A great deal of the breakup happens in the phase inverter (which most people enjoy), and the master volume reduces some of that breakup (which is why people didn't like this type of master volume in this circuit).

The input cathode cap is large on the normal channel, but you won't notice much of a reduction in bass content until you reduce it to about 3.3-4.7μF. You can try reducing this to customize the sound, but I think you're better off just running that channel at lower volume and getting much of the sound from the lead channel

Beyond that, you can absolutely modify the circuit to your liking.  However I think you're better off with another circuit like an AB763 type.

[uki]

The way it works is indeed very interesting, yet the amp does break up too much for my style, I'm used to the clean and big headroom of Fender like amps.

Beyond that, you can absolutely modify the circuit to your liking.  However I think you're better off with another circuit like an AB763 type.

You are probably right, that was my second though to modify the preamp to some Fender circuit. I'm doing it,  i thinking a twin like , have any suggestions ?!

Above all I really appreciate the time and effort helping me, the fact that super early drive been solved is a game change, thanks a lot !! 

[stratomaster]

The pre phase inverter master volume is my recommendation. Try that before anything else. You can even use one of the pots you removed from the amp earlier to quickly test if you like what it does.

You can also try making R16 47k-56k and leaving the coupling caps reversed as they are. This will shift the eq of the amp more towards treble than bass and low mids. This might be the clarity and bass reduction you're looking for.

[uki]

Tried the 56k I did shift the eq a bit the normal channel is more pleasant now, but still mudy very low.
I'm starting to realize this circuit is not for me. Too much grit and break up.

Sounds like a good idea to change the preamp to a Fender twin like. i got a couple of concentric pots for the tone stack, so avoiding to drill more holes in the chassis.

I got this schematic I've modified while ago, its just a draft, values may be wrong, and some more tweaking is probably needed, also mid control missing in one side,  tell me what you think:

[stratomaster]

What taper are you using on your bass pot? My experience with the circuit is the lead channel is very bright and brash and the wooly warmth of the normal channel is a welcome contrast and very useful for blending--again at very low levels.

The plexi in stock form is capable of beautiful clean tones with some careful tweaking, especially jumpered.  It's just a few components different from a tweed Bassman after all. 

Try the master volume before completely redesigning the amp.

You can put a 2.2μF on the normal channel cathode to decrease the low end further or reduce the coupling cap even more.

You can replace the 470k mixers with 220k-270k for lossier mixing without abandoning vintage inspired values.  This will affect the bright cap performance.

You can add 22k resistors to the plates at the socket for a split cathode gain reduction.

Lots you can do before a major redesign.

[uki]

What taper are you using on your bass pot?

1mA alpha  one side read 142 and the other 926, across the pot 1064

Try the master volume before completely redesigning the amp.

Have not tried that yet, I dont want to add another pot to the chassis, the faces finishing is very fine.

You can put a 2.2μF on the normal channel cathode to decrease the low end further

Done that with good result, normal is now way more pleasant than the original setting.

reduce the coupling cap even more

I does feel like could change that a little, to what value thou?

You can replace the 470k mixers with 220k-270k for lossier mixing without abandoning vintage inspired values.

Done, the two resistor in the amp have drifted in value, they were close to 470k back when the amp was built,
now they read 487k and 509k.
I found two resistor reading 255 ish, now in place (note: the 2.2m resistor still in place)

This will affect the bright cap performance.

It did, the bright channel got sharper now.

You can add 22k resistors to the plates at the socket for a split cathode gain reduction.

I didnt understand how to do it.

I got a questions:
The resistors in the PA tubes(R31,R32) screen were originally 1k and were replaced with 2.2k in one of the atempt to reduce the distortion, should go back to original value or what value would best fit for a clean tone?

The OT is ultralinear !! Could those taps be used for cleaner tone ?

[stratomaster]

If you like the normal channel now, then there's no need to reduce the coupling cap further.  Next standard size down is 0.001μF. This would be pretty unusual for an input stage coupling cap on a clean voiced amp.

You really should try the master volume. You can just temporarily install it for testing purposes.  If you like it, then find a place for it on the back of the chassis or even on the bottom. Or if you find a particular setting you like you can used fixed resistors hardwired after the treble pot to create the same voltage divider.

As far as the split plate load, you just disconnect the wire to the plate at the tube socket and add the resistor there just like you would for a grid stopper. You're just squeezing a resistor in between the wire and the socket. 22k might be a bit high, but you can experiment once you get the idea.

I doubt you'll hear a big difference between the 1k and 2.2k screen. I'd leave this as is for now.

Leave the transformer as wired.  Most of the breakup is in the PI--which is why I'm so adamant about trying the master volume.  You can get the amp to have a but more output power, but to reduce the breakup/distortion the MV is really the direct approach.

[uki]

I doubt you'll hear a big difference between the 1k and 2.2k screen. I'd leave this as is for now.

What is the range for the resistor?

Most of the breakup is in the PI--which is why I'm so adamant about trying the master volume.  You can get the amp to have a but more output power, but to reduce the breakup/distortion the MV is really the direct approach.

The breakup still too early about 5-6 volume, ill try the volume thing.
Does the position of the presence control have anything to do with it? Its in the PI

When turning up the volume at 4 and above the amp sounds like a person trying to speak with mouth full of food,
that is what i mean by clarity, it sounds very grainy

I did read about the V2 cathode follower been a source of the distortion, what can you say about it?
bring me back to the idea of the modded schematic, tone stack between V2 triodes follow the schematic where i got the idea

Bright channel is too bright very sharp after the mix resistors change, harsh on ears, im guessing changing the .022 C3 cap would bring it down ?

I'll try the plate resistors, which plates?

[stratomaster]

I doubt you'll hear a big difference between the 1k and 2.2k screen. I'd leave this as is for now.

What is the range for the resistor?

There is no range for leaving it alone. 1k is a popular value for EL34 screens.  2.2k will give the screen a good bit more protection and result in a touch more compression and slightly lower power. You may be able to measure this, but I doubt you'll hear it.

Most of the breakup is in the PI--which is why I'm so adamant about trying the master volume.  You can get the amp to have a but more output power, but to reduce the breakup/distortion the MV is really the direct approach.

The breakup still too early about 5-6 volume, ill try the volume thing.
Does the position of the present control have anything to do with it? Its in the PI

The presence control is in the negative feedback loop, and it absolutely affects breakup.  Play around with the settings and how it interacts with the treble and volume controls.  You can get cleaner sound by turning the presence down and treble up.   

When turning up the volume at 4 and above the amp sounds like a person trying to speak with mouth full of food,
that is what i mean by clarity, it sounds very grainy

I did read about the V2 cathode follower been a source of the distortion, what can you say about it?

It has been shown to be a source of compression and it reduces the insertion losses of the tonestack.  You can try bypassing it for experimentation purposes, but again you're looking everywhere but where the breakup you don't like is happening. Put in the pre PI master.

bring me back to the idea of the modded schematic, tone stack between V2 triodes follow the schematic where i got the idea

Bright channel is too bright very sharp after the mix resistors change, harsh on ears, im guessing changing the .022 C3 cap would bring it down ?

No. Coupling caps act like a high pass filter on combination with the plate impedance.  Coupling caps determine where the bass roll off occurs and do not directly affect the high end (ignoring phase effects).  I'd be inclined to just remove the bright cap and try it that way. Removing the bright cap on a plexi is a popular mod.

I'll try the plate resistors, which plates?

Any stages where you want to reduce gain.  I'd do the stage before the cathode follower first because it affects both channels with less work.  For even less work and more gain reduction you can try a 12AU7 in V2.

[uki]

The resistor to the plates of V1 did work well with the bright channel, volume goes up to about 6 no break up,
 22k feels like is a bit too much, ill try bring it down a bit, 18k maybe 15k

not so much with the normal, it still mudy. cathode cap change maybe give it a bit more life?

The master volume did about the same as the plate resistor, tested with both together and isolated.
It does clear up the tone further, i rather keep the MV out thou.

Question about NFB , as far as i know changing it can give more headroom and clean tone, can that be done?