Trainwreck Rocket AC30-ish build NOW WITH REVERB

Keefy

Starting to look a bit more like the amp it's going to be:

Keefy

Wiring up in progress. Only an idiot would install 1W 100k resistors where the  5W 100ohm screen resistors should go - at least I noticed early on.

Also, Creamback Neo installed in cab.

Keefy

Added some wires to the board:

Keefy

Board installed and wired to input jack and pots. No sense in rushing this job!

PeteC

Very nice work indeed ! - looking a lot neater than my express build Graeme.  Going to be an amazing amp.   Ive settled on 6V6s in the Express for the time being and just finished the Airbrake attenuator yesterday which means I can now crank it up in the house a bit.  Worth considering if you need to knock the volume back a bit when its built. 

Keefy

PeteC said:

Very nice work indeed ! - looking a lot neater than my express build Graeme.  Going to be an amazing amp.   Ive settled on 6V6s in the Express for the time being and just finished the Airbrake attenuator yesterday which means I can now crank it up in the house a bit.  Worth considering if you need to knock the volume back a bit when its built. 

Thank you

Keefy

I think that's all the wiring except the mains and the heaters. First I'll go through and check connections and continuity.

The observant will note no connections to V4 and to half of V1 - these are reserved for a future reverb circuit.

Keefy

It lives! It's hot!

I finished all the wiring this morning and checked connections and continuity. Then using a variac and a dim bulb tester, I checked the operation of the heater circuit, then the whole thing without valves, then with preamp valves. I currently have it on my bench, wired up to an unboxed 10in speaker.



So far I have identified 4 issues:

1. It runs very hot - readings below - but I can't see red plating.
2. There is a lot of hum, even with the volume control at zero.
3. When I jiggle the cathode connection from the board to V2 (V1a in the circuit diagram, which comes after the volume control - second short yellow wire from the right in the photeo), the hum disappears, but so does all the signal.
4. The treble and bass controls don't necessarily do quite what you expect. They are more interactive than the ones on my Matchless Lightning! This may not be a fault of course.

All readings taken at 249V mains, all voltages referenced to earth:

PT secondary 268-0-268 VAC
Heaters 3.08VAC
Power draw at standby 35W
Power draw, on and idling 114W (seems about right)
OT resistance to CT(red) : brown side 84.3R, blue side 93.6R

B+1 341V
B+2 330V
B+3 235V
B+4 201V
B+5 197V

Brown side plates 333V
Blue side plates 330V

Cathodes 10.32V

Brown side
Plate current = (341V-333V)/84.3R = 94.9mA => 47.4mA per valve
Plate dissipation = (333V-10.32V)x0.0474A = 15.3W

Blue side
Plate current = (341V-330V)/93.6R = 117.5mA => 58.8mA per valve
Plate dissipation = (330V-10.32V)x0.0588A = 18.8W

That seems a lot for a valve with an 'official' maximum PD of 12W, but according to Ampbooks the AC30 Silver Jubilee model runs at 17W. I should be able to reduce it by fitting a larger value of cathode resistor.

The one thing I've done differently from the Modulus layout is to make the preamp earth connections via the input socket only, and not a separate chassis grounding bolt. This is something Rob Robinette recommends on his 5E3 layouts - I might try changing that back.

I'm going to give this a rest for today. Tomorrow I will come back to it, troubleshoot the issue around V2, and maybe add another earth runner (I already drilled a hole in the right place when I was at the metalwork stage). I'll also order a selection of 25W resistors.

ICBM

I think there may be an error in your readings if the current draw on each side of the power amp is that different - even with a badly mismatched set of valves that's more difference than I would expect. Try swapping one of the 'hot side' valves with one of the 'cold side' and see if anything changes. AC30s do run the valves at about 16W and don't go through them *too* fast so about that is not ridiculously hot.

It sounds like there's a bad contact on the cathode for V2, but if the hum disappears when the signal does, at least you know it's coming from the early stages of the preamp.

I forgot to mention earlier that it's a good idea to connect all three terminals of the unused side of V1 to ground, until you use it for the reverb - I doubt that's the cause of the hum, but it's not good for the valve to have the filament hot and the electrodes floating at random charges.

Neat work .

Keefy

I assumed the mismatch between the two sides of the OT primary was mainly to blame. I don’t think Carvin went all out for quality when they were sourcing components! FWIW my Orange valve tester tells me that all the EL84’s are pretty well matched.

 I will ground those unused terminals when I look at the amp tomorrow.

Thanks for your kind comment. I don’t subscribe to the practice of laying wires out rigidly to straight lines and right angles, but a tidy layout makes everything that much easier.

PeteC

Nearly there.... definitely resolder the connection to the cathode of v2. Where the cap and resistor connect to your turret on the  board.  Looking at the pics quite a few connections to the terminals look like they need more solder to ensure a really good joint.  I use eyelets now but have found with turrets that you can need a lot of heat to get a very good connection. Often they can look absolutely fine at first glance , and may seem mechanically connected , but are not actually well soldered.   I would try reflowing them and seeing if that helps.  

Good luck.  

Keefy

PeteC said:

Nearly there.... definitely resolder the connection to the cathode of v2. Where the cap and resistor connect to your turret on the  board.  Looking at the pics quite a few connections to the terminals look like they need more solder to ensure a really good joint.  I use eyelets now but have found with turrets that you can need a lot of heat to get a very good connection. Often they can look absolutely fine at first glance , and may seem mechanically connected , but are not actually well soldered.   I would try reflowing them and seeing if that helps.  

Good luck.  

Cheers @PeteC ;. On this build I tried to use enough solder to make a decent connection where leads go down inside the turrets, without necessarily filling them. I think I'll go back and fill 'em all up, as you can't actually see inside to see how good the connection is.

PeteC

You probably know this but ....Turrets can act like a heat sink unless you have a good high powered and temperature controlled iron to stop the tip heat dropping suddenly when you contact the turret.  After a few builds where I was going for the “turret look” I reverted to eyelets to be sure of a good solder joint without having to heat the turret and the component leads for too long.    Worth reflowing them to be sure of a good electrical joint in case any look Ok but are actually “dry”. 

Thrilled with sound of the express build which I’ve been playing through tonight - but god it’s loud , even with the airbrake on most settings.   Expect a loud amp !  Does the Liverpool have a master - I can’t remember without looking at the schem.?  

Keefy

PeteC said:

You probably know this but ....Turrets can act like a heat sink unless you have a good high powered and temperature controlled iron to stop the tip heat dropping suddenly when you contact the turret.  After a few builds where I was going for the “turret look” I reverted to eyelets to be sure of a good solder joint without having to heat the turret and the component leads for too long.    Worth reflowing them to be sure of a good electrical joint in case any look Ok but are actually “dry”. 

Thrilled with sound of the express build which I’ve been playing through tonight - but god it’s loud , even with the airbrake on most settings.   Expect a loud amp !  Does the Liverpool have a master - I can’t remember without looking at the schem.?  

I have a 60W temperature-controlled iron. As long as I can get good thermal contact with the turret then I can heat it up reasonably quickly. I heatsink the components with croc clips.

No master volume on the Rocket! Master volumes are for wimps

Keefy

This afternoon I:

• Re-flowed any joints that seemed to be low on solder
• Added a (temporary) ground runner to unused pins 1 2 and 3 on V1
• Added a chassis earth to the pre-amp (in addition to the input jack)

Switched on - no hum, but no signal either.

I began to suspect a bad valve but testing showed them all to be functional. Even so, swapped out V3 (no-name ECC83) and replaced with a Shuguang.

Checked all resistances to earth - all correct. Checked continuity to all B+ nodes - all good.

Switched on - hum is back, but so is signal! After playing for a couple of minutes, the sound died away over a few seconds like when you switch off.

Checked all heater voltages - fine.

Checked B+ nodes - aha! Here are the differences:

Node	Voltage when working	Voltage when no sound
B+1	341	341
B+2	330	332
B+3	235	303
B+4	201	303
B+5	197	299

This is making me think that the earth connection is intermittent in the later stages of the power supply, although it all tests out fine for continuity.

Also I'm pretty sure the hum is 100Hz, so not the heaters (which would be 50Hz).

Enough for today, I'll come back to it tomorrow.

ICBM

That looks like no current is being drawn through the preamp valves - which allows the voltage at the far end of the B+ chain to rise almost to the supply voltage - if it fades in and out, you've got a bad connection in the filament supply.

PhilKing

I  noticed that on V4, the black heater wires got to separate tags, if there is no valve in this, could that be the problem (I'm assuming that the two tags is standard for the two halves of the valves).  I'm not a technical expert on this, but just wondered if that might be the issue, especially since it seems to be the heater voltage that's the problem

Keefy

ICBM said:

That looks like no current is being drawn through the preamp valves - which allows the voltage at the far end of the B+ chain to rise almost to the supply voltage - if it fades in and out, you've got a bad connection in the filament supply.

PhilKing said:

I  noticed that on V4, the black heater wires got to separate tags, if there is no valve in this, could that be the problem (I'm assuming that the two tags is standard for the two halves of the valves).  I'm not a technical expert on this, but just wondered if that might be the issue, especially since it seems to be the heater voltage that's the problem

Thanks chaps, you were both on the money as regards the loss of signal.

I put the amp into standby with the shielding cans removed so that I could see all the filaments - so far. so good. Then I switched in the HT and heard the hum, but after a few seconds it faded away and there was no signal. Looking at the valves, V1 and V2 had gone dark, but V3 was still glowing.

When I wired up the filaments, I connected pins 4 and 5 on the preamp valves with a short wire link. I then daisy chained the supply in on pin 4 and out on pin 5 of each valve (easier than trying to fit 3 conductors into one hole on a valve socket lug). Probing with my meter found no heater voltage on pins 4 and 5 of V1 and V2. Examining the connections with a magnifying glass, I found the wire link on V3 had escaped soldering, so it was sometimes making an adequate connection, sometimes not.

That's now fixed - thanks chaps

Next up is the hum. The Modulus layout has both PT secondary centre taps (I think - it doesn't actually say) joining the board at the negative end of the first filter cap. This point is the centre of a star wiring scheme, with all the other filter cap negatives running to it. The input socket ground runs to the negative of the last filter cap, as does a runner to the chassis. It's not clear where the cathode resistor for the EL84s gets its ground.

I have gone for a different approach: The PT CTs and a runner form the first three filter caps go to the chassis, as does the cathode resistor. The last 2 filter caps and the input socket are grounded as per Modulus. Maybe I was being too clever...

I am going to try re-routing the earths in line with the Modulus scheme - I'll let you know how I get on!

Keefy

I had high hopes for the earthing scheme, but it made not one bit of difference. Neither did moving the amp away form all the other electrical devices in my workspace.

Next I looked at lead dress. With the amp on, I poked at various flying leads with a chopstick. By moving the wires that run to the PI grids closer to the chassis, I was able to reduce the hum by about half. The OT primary leads run most of the length of the circuit board (twisted tightly together) and pass under the PI grid leads. Pushing on them with a chopstick also affects the hum. I think I may re-site them to the underside of the chassis and bring them back through to where they connect to the power valves. I can then look again at the siting of the flying leads.

ICBM

If the hum went away when the filament connection was broken, it’s definitely coming from the preamp rather than the PI or later.