Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: Fiat_cc on November 13, 2019, 08:23:23 pm
-
I’m beginning to lose faith. Everything I build lately has problems, and it is driving me nuts. Anyway.
I have a build that is more or less 6G6A guitar channel, and aa864 bass channel (although the tone stack is closer to ab165), running through a Twin 100W output section.
I am having trouble with farting out, and nasty ratty sounding distortion. I also have some
Hum trouble with the bass channel. I can’t post schematic or layout of exactly what I’ve built as my laptop got stolen a couple of weeks ago, and I hadn’t properly backed up.
Anyway...
Both channels exhibit the farting out tendency. It is more noticeable with bass, but it does it with guitar too. It sounds almost like a harmonic an octave below.
I’ve put 100Hz in to it through a signal generator, and scoped it (the bass channel at least, I haven’t done the guitar channel yet). The distortion appears to happen at the phase inverter. Signal is clean up to there, then gets nasty. It happens at relatively low settings on the volume pot too with a 200mV p-p 100Hz input.
Component values and wiring appear to be correct. Could it be a parasitic oscillation of some sort? I know that the amp will overdrive, but this doesn’t sound like nice smooth overdrive, and it happens while the rest of the preamp is still very clean. If you push further, you can hear the onset of normal, pleasant distortion.
I’m not sure what to look for with this one.
-
If you want help then please post Hi Res pics and the schematics you used. Jim
-
If you want help then please post Hi Res pics and the schematics you used. Jim
Hi Jim.
As I mentioned in my post, my laptop has been stolen. I don’t have the schematic or layout drawings, and for some stupid reason, I can’t post the chassis pics off my iPhone because the jpeg file extension is wrong and the forum won’t accept them. I can try to do a pencil drawing of the schematic later, but until my new computer arrives, that’s about all I can do.
I’ll link to the schematics I used for reference though if that helps.
http://ampwares.com/schematics/bassman_6g6a.pdf
http://ampwares.com/schematics/twin_reverb_ab763.pdf
http://www.ampwares.com/schematics/bassman_aa864.pdf
http://fenderguru.com/wp-content/uploads/2012/08/fender_bassman_ab165-fenderguru-mods.gif
So, the guitar channel is the guitar channel from the 6G6A.
Bass channel is bass channel from aa864, with tone stack from ab165 with an added mid control.
Output section is Twin.
Each channel has a coupling cap on the plate, feeding a 220k mix resistor, into a 500pF coupling cap into the PI (like the Twin). I started with a higher value here, but reduced it to try and reduce the bass farting out effect. It made minimal difference.
The distortion is occurring in the phase inverter according to injected signal and scope.
-
Could OT secondaries running too close to the phase inverter wiring cause a parasitic oscillation?
-
Disconnect the NFB wire. Still do it?
-
Disconnect the NFB wire. Still do it?
I need to play through it, but with signal injected from a signal generator running in to dummy load, and looking on the scope, lifting the negative feedback cleans up the wave form massively. It looks as though there is more amplitude on the signal too, so I’m not sure what’s going on. I’d have expected if I was injecting positive feedback that a) it would squeal and carry on like other amps where I’ve gotten the phase reversed, and b) disconnecting it would reduce the amplitude of the signal, not increase it.
-
Actually, just compared photos of the scope traces. Amplitude is almost identical, but with NFB, the wave form is more distorted.
-
However, playing bass through it suggests that NFB isn’t causing the issue. I still have the distortion/ghost harmonics thing going on.
-
The best way I can describe the sound is like when you play a chord with a really wrong note in it and you get unpleasant bump frequencies interacting, only it’ll do it on single notes, and on chords that are in tune and correct.
It’s gotta be an oscillation of sorts, but what I also found playing around with bass and signal generator through a speaker, is that the distortion is very audibly apparent, with no real visible issue on the scope trace.
-
Could OT secondaries running too close to the phase inverter wiring cause a parasitic oscillation?
I’m curious about this. I currently have the OT in roughy the centre of the chassis, and both primary and secondary OT leads enter the chassis directly under the turret board. The primaries are basically directly below the phase inverter circuitry, and the secondaries are closer to the final gain stage of the bass channel. It is a monumental pain in the a** to move them just as an experiment that doesn’t work (would basically require drilling holes in the chassis), but I’m running out of other ideas.
Fender brings these leads into the chassis much further away from the eyelet board, and I am assuming for good reason. This could be a big oversight in my chassis design.
-
the distortion is very audibly apparent, with no real visible issue on the scope trace.
mic the speaker into the scope, or swap speaker. OT wires carry massive AC on the primaries so lead dress is important.
scope the PS rail, look for a 5-20hz slow roller on top of the VDC
-
On a totally unrelated side note: Do you use flux on your solder joints?
-
Disconnect the NFB wire. Still do it?
I need to play through it, but with signal injected from a signal generator running in to dummy load, and looking on the scope, lifting the negative feedback cleans up the wave form massively. It looks as though there is more amplitude on the signal too, so I’m not sure what’s going on. I’d have expected if I was injecting positive feedback that a) it would squeal and carry on like other amps where I’ve gotten the phase reversed, and b) disconnecting it would reduce the amplitude of the signal, not increase it.
They don't always squeal. Swap the OT primary plate leads to determine if you have the phase wrong.
-
Have you looked at the schematics for the CBS Bassman 100 which has the quad 6L6 output circuit. It may be that one or more of the values in the PI or in coupling need to be changed. Compare that to the Blond Twin (Showman) circuits - 6G14/6G14a
I know that putting early circuits with later circuits sometimes doesn't work unless all of the small nuances are checked.
And sometimes you get Frankenstein.
You might need to get one channel sorted out and then the other.
I'd pull up the classic "blackface twin" layout (or get the one from Mojotone) and study the quad output wiring to see how Fender got it right. I've worked on a lot of Twins with problems but never any with the output section (except the one with a beer spilled in it) so Fender got something right.
good luck
-
On a totally unrelated side note: Do you use flux on your solder joints?
Sometimes yes, sometimes no. These particular Keystone turrets flow solder well without, but the cheap Chinese turrets I had before these required flux use.
-
This was a kind of death by 1000 cuts situation. I ended up being dissatisfied with the tone of the amp too.
In the end, I moved a few grounding points, re-routed the output transformer leads outside the chassis for longer (a la Fender Blackface style), changed the input to the phase inverter to 470pF, changed the outputs of the phase inverter to 0.022uF, and touched up the lead dress. Hum is now more than acceptable, and it doesn’t fart out like it did.
All in all, I’m now happy with it.