Trainwreck Rocket AC30-ish build NOW WITH REVERB

Keefy

It wasn't the B+3/B+4 link wire.

I'm now back to thinking it's a ground scheme issue. If I ground the output of the tone stack (R10/C8 junction) across the 220k resistor that runs from that point to earth (not in the original schematic), the hum goes. If I ground it to the negatives of filter caps 1 to 3, hum goes. If I ground it to the far end of the wire that runs just 4 inches from those filter caps to the chassis, loud buzz. If I ground it to the negatives of filter caps 4 or 5, or to the input jack, loud buzz.

Come the weekend I'm going to completely re-do the grounds, this time following the Modulus layout. I'll also be reinstating B+4 and B+5 of course!

In related news, I replaced the chunky 50R cathode resistor with a 68R. Plate dissipations are now about 12W and 14W - less than in an AC30, but not exactly biased cold! I am happy that my EL84s are not being cooked to death. I can always tweak this up a bit at my leisure.

PeteC

hope the regrounding gets it sorted for you - I know how frustrating it can be when you have done everything per the right layout and it still won't function well. 
I'm sure your close to having it done. 

ICBM

Keefy said:

In related news, I replaced the chunky 50R cathode resistor with a 68R. Plate dissipations are now about 12W and 14W - less than in an AC30, but not exactly biased cold! I am happy that my EL84s are not being cooked to death. I can always tweak this up a bit at my leisure.

That's a common mod for AC30s if they redplate as well.

Are you sure about the resistance readings for both sides of the OT? They're never identical - the outer of the two halves is always higher, because there has to be a longer length of wire for the same number of turns - but the difference shouldn't be enough to produce that much difference in the valve dissipation normally.

Keefy

PeteC said:

hope the regrounding gets it sorted for you - I know how frustrating it can be when you have done everything per the right layout and it still won't function well. 
I'm sure your close to having it done. 

Cheers! I had made a small grounding mod that worked on a 5E3 build, but it's clearly not right as I have it at present.

ICBM said:
...

Are you sure about the resistance readings for both sides of the OT? They're never identical - the outer of the two halves is always higher, because there has to be a longer length of wire for the same number of turns - but the difference shouldn't be enough to produce that much difference in the valve dissipation normally.

Yes, 100% sure on the OT resistances. I haven't tried swapping any EL84s around yet but their 'matching values' (courtesy of my Orange Valve Tester) are close - three 8 scores and one 9. They are all JJs that have done a year's gigs in my Matchless.

Keefy

Since my last post I have been reading up on grounding schemes. I read the Valve Wizard chapter on grounding and realised I had made the error of 'using the chassis as a more-or-less random ground’. To quote Mr Blencowe, 'Allowing it to become part of the audio circuit by treating it like one big fat wire is asking for trouble.'

The donor Carvin amp had the PT secondary's centre tap grounded at the chassis. I had kept this and also grounded the output valves' cathode resistor there - probably wasn't the best idea.

This morning I completely re-did the grounding to follow the Modulus scheme, which I now realise is a 'multiple star' layout.

There is still hum, probably slightly reduced, but it's no longer influenced by the position of the treble control. It still goes completely when I earth the tone stack output (R10/C8 junction).

Looking at what feeds the tone stack, I measured V2a's cathode at about 128V DC - is that within normal range?

I did a bit more poking with the chopstick and it makes some difference where I push the OT primary wires. I previously tried running them outside the chassis (in an insulating sleeve) and it made no difference, but that was before I re-did the grounding scheme - I might try this again, as they do run very close under the board. I'm a bit limited for options on this because of the topology of the donor chassis.

ICBM

Keefy said:

Looking at what feeds the tone stack, I measured V2a's cathode at about 128V DC - is that within normal range?

Yes - it's a cathode-follower position. This is why you should not use a Russian or JJ valve there, they have poor cathode-to-filament insulation and can fail. The spec for the 12AX7 says the rating is 180V so it should be fine, but modern valves clearly aren't all made to the proper spec.

Keefy

ICBM said:

Keefy said:

Looking at what feeds the tone stack, I measured V2a's cathode at about 128V DC - is that within normal range?

Yes - it's a cathode-follower position. This is why you should not use a Russian or JJ valve there, they have poor cathode-to-filament insulation and can fail. The spec for the 12AX7 says the rating is 180V so it should be fine, but modern valves clearly aren't all made to the proper spec.

Thanks, that makes sense about he cathode follower. The valve is a Shuguang btw,

Keefy

I came back to this yesterday to see what I could do. I git rid of a little bit of noise by moving the OT primary cables away from the board. Going back to the circuit diagram I can see that Modulus's additional 220k resistor from R10/C8 to ground has the effect of attenuating the output from the tone stack by about 50% (R10 is also 220k). I replaced R10 with 500k (2x 1M in parallel as I didn't have a suitable 470k!) which resulted in a reduction of both guitar level and hum through my bench speaker.

I then went to put the chassis in the cab, only to find I had cut the front panel face plate about 1mm too large to fit! This morning I dug out my cheap-ass wood plane and carefully shaved down the projecting lip of plastic. I installed the chassis, plugged in my Strat, and fired it up.

The good stuff
- It sounds pretty good. The odd effect of my free-floating bench test speaker has been replaced by a full and fruity sound.
- The tone controls work really well, with plenty of range.
- The cosmetics look right, allowing for the somewhat road-worn tweed!

The not-so-good stuff
- It still hums at a level that annoys me.
- Having knocked back the level going to the PI am not sure this will give me 30W of loud. I had the volume up to 1 o'clock, which is as loud as I dared in a terraced house. I would really need to hear this in a band context.
- There is a stupidly loud rattle in response to low notes, probably from the valve retainer springs. I might try removing them.
- The chassis sits further back in the cabinet than I anticipated, meaning the mains power switch sticks out past the outline of the cab. You couldn't lie the amp on its back to transport it without risking damage. It wouldn't be too difficult to install the mains rocker switch next to the standby on the front panel - the cutout is still there under the face plate.
- Also the speaker jack sticks out a bit too much, but that could be recessed by fitting couple of washers between the socket and the inside of the chassis

So where next? For as long as this amp hums like it does, I am never going to love it. If I can't get shot of the hum then I will part it out and do a different build somewhere along the line. There's nothing in it that can't be re-used or re-purposed for another project.

If I can fix the hum then it will be worth my while addressing the other issues, and possibly adding spring reverb. Two more things occurred to me about the hum:
- If I moved the volume control to just before the PI this might increase the signal-to-noise ratio.
- The whole chassis (which is 14-gauge steel) can be felt gently vibrating - presumably the PT causes this. Could this be inducing hum into the tone stack components?

PeteC

Have you already eliminated the heaters as a possible source of hum ( lead dress etc ). I like Merlin Blencowes method for heaters.  It seems to reduce hum levels for me.  

 Is it worth elevating the Heaters to see if that helps ? Might be straightforward in a cathode biased amp taking a dc feed for the heater Centre Tap from the cathodes of the output stage ?  But check and read up first !!   I am sure someone more knowledgeable will advise    I am not that familiar with the amps topology so only given some rough ideas here.  

Also from what you describe the PT might be a problem if there is that much vibration going on.  

Keefy

PeteC said:

Have you already eliminated the heaters as a possible source of hum ( lead dress etc ). I like Merlin Blencowes method for heaters.  It seems to reduce hum levels for me.  

 Is it worth elevating the Heaters to see if that helps ? Might be straightforward in a cathode biased amp taking a dc feed for the heater Centre Tap from the cathodes of the output stage ?  But check and read up first !!   I am sure someone more knowledgeable will advise    I am not that familiar with the amps topology so only given some rough ideas here.  

Also from what you describe the PT might be a problem if there is that much vibration going on.  

With V2 (which immediately precedes the tone stack) pulled, the hum persists, but if I ground the output of the tone stack (before it enters V3) it goes. It is coming from the tone stack, not the heaters. I have run the heater wires up, across, and down to the next valve, keeping them twisted for all but the last 1cm of each run, and away from signal wires. They run to the pilot lamp first, then the output valves, then the preamp from V3 to V1. I can't elevate the heater voltage, because they share the same centre tap as the PT secondary via an internal connection.

As to the vibration, I get no detectable vibration from the chassis of any of my other amps.

ICBM

It does sound like hum being induced into the chassis is a possibility. Try running it with the transformer unbolted from the chassis and held up on small blocks of wood or something - obviously the chassis will have to be upside-down to allow this.

Keefy

ICBM said:

It does sound like hum being induced into the chassis is a possibility. Try running it with the transformer unbolted from the chassis and held up on small blocks of wood or something - obviously the chassis will have to be upside-down to allow this.

Cheers, will try that

PeteC

looks like you have enough space to use another small TX just for the heaters as well if you wanted to try DC heaters or a DC elevated heater circuit.  Mind you, increasingly sound like a problem mains TX to me, or something odd in the grounding around your tone stack. 

hope you get it sorted soon - lovely job and a lot of work you have put in so far. 

Keefy

PeteC said:

looks like you have enough space to use another small TX just for the heaters as well if you wanted to try DC heaters or a DC elevated heater circuit.  Mind you, increasingly sound like a problem mains TX to me, or something odd in the grounding around your tone stack. 

hope you get it sorted soon - lovely job and a lot of work you have put in so far. 

The tone stack is grounded as per the Modulus layout (as is everything else now).

I tried unbolting the PT while supporting it from beneath. Mechanical vibration of the chassis disappeared, but didn't seem to make a lot of difference to the hum so I put it back. Not convinced the hum is as bad as it was, but have no idea why that might be!

There's a 1970s AC30 PT on eBay at present (no 5V winding, but I don't need that) but looks like it has a loose lamination, which could just mean a different lot of trouble:

PeteC

If modulus don't have the Trainwreck type mains TX in stock at the moment you could also try the Hammond 372JX maybe ?  What current draw is the Liverpool output stage ?     the Hammond is about 290ma and under-rated 

Keefy

I decided to tackle some of the issues that I found when I first ran the amp in combo format.

The mains power is now a rocker switch on the chassis front panel next to the standby switch. The speaker output jack is now mounted flush so that the plug does not project beyond the cab profile. I also reversed the resistor mod thatI had done. So far, so good.

Next the rattle, which sounded like it could be coming from the EL84 retainer springs. When I installed these there were no wires to get in the way - that has changed somewhat! Rather than remove them, I opened up some insulation sleeving and heat-shrink tube, and popped a layer of each around each spring.

Firing up again inside the combo, the hum is still louder than I would want, and the bloody rattle is still there. However, with a Les Paul this amp is a beast. The amp breaks up in a very satisfying way, starting with that clean-but-dirty sound that can be so hard to achieve. And it's loud. I haven't dared to get it up to Brain May levels yet.

I'm going to have another go at running the OT primary leads outside the chassis and therefore away from the board . When I first did this, there was still an issue with my grounding scheme in that I hadn't followed the recommended one! If that doesn't work, I really am stumped. I don't want to spend a chunk on a new/replacement PT when there is no guarantee that it will fix the hum problem.

PeteC

Often hard to pin down rattles are coming from the tubes themselves.   Sometimes tubes sold as new high quality valves can be quite rattly.   Have you tried running the chassis well away from the speaker to eliminate that ?  Swapping out tubes one at a time can eliminate the culprit. Also worth hot gluing down capacitors to the board sometimes as these also resonate microphonically especially in combos near a speaker.   Not just ceramics.     It still sounds to me though like something is up with that power tx - or it’s not up to the job.   

Regarding the hum - can you measure the ripple at each stage of the power supply - perhaps someone on ampgarage can compare with their own Liverpool build ??     

Keefy

PeteC said:

Often hard to pin down rattles are coming from the tubes themselves.   Sometimes tubes sold as new high quality valves can be quite rattly.   Have you tried running the chassis well away from the speaker to eliminate that ?  Swapping out tubes one at a time can eliminate the culprit. Also worth hot gluing down capacitors to the board sometimes as these also resonate microphonically especially in combos near a speaker.   Not just ceramics.     It still sounds to me though like something is up with that power tx - or it’s not up to the job.   

Regarding the hum - can you measure the ripple at each stage of the power supply - perhaps someone on ampgarage can compare with their own Liverpool build ??     

I might try it with the chassis on blocks on top of the cab so that I can prod possible culprits. It does sound like the kind of noise a spring would make.

I have an old moving coil meter with a setting for measuring AC in the presence of DC so I'll try measuring that.

I think you may be right about the PT as it seems a lot smaller than the Modulus AC30 PT, even allowing for lack of a 5V secondary.

PeteC

I would run a long cable to the speaker and get the chassis well away from any mechanical vibration to be sure. 

Keefy

PeteC said:

I would run a long cable to the speaker and get the chassis well away from any mechanical vibration to be sure. 

There's no rattle when the chassis is supported on foam blocks on my worktop next to my bench speaker.