Trainwreck Rocket AC30-ish build NOW WITH REVERB

Keefy

I discovered that most of that hum was coming from the reverb tank itself. On further investigation, this was because it was only a foot or so from my Variac! With the Variac out of the equation the reverb is hum-free until the level control gets past 3 o'clock.

I've tried two tanks, a Mox short-tank 3-spring long decay, and an OC Electronics Folded Line. Both sound fine, but I still have to turn up the level control further than I would on a Fender amp. The amount of reverb available certainly doesn't reach surf levels! I wonder whether the driver could be made to push the tank harder, for more 'dwell' I suppose.

Here's what the reverb driver circuit looks like now:

ICBM

Keefy said:

I calculated that the two triodes in the 12AT7 driver are only dissipating 0.54W between them (as against an absolute maximum of 5W) so it's not going to be red-plating any time soon!

Ah... I may have miscalculated. The voltage drop across the 47K plate resistor will be much greater than the OT primary in a Fender-type circuit. Probably worth experimenting with lower values - at a guess you want about 2W total for both plates or the reverb might sound a bit harsh because the valve is cutting off some of the waveform.

Keefy said:

There is now plenty of reverb available. There is a noticeable increase in background hum as the reverb level is raised. I will spend some time poking wires with a chopstick to see if I can reduce this with lead dress.

Try changing the orientation of the reverb tank, it's most likely just hum pickup from the return transducer. Make sure the Output end is at the opposite end of the amp from the power transformer first...

Keefy

ICBM said:

Keefy said:

I calculated that the two triodes in the 12AT7 driver are only dissipating 0.54W between them (as against an absolute maximum of 5W) so it's not going to be red-plating any time soon!

Ah... I may have miscalculated. The voltage drop across the 47K plate resistor will be much greater than the OT primary in a Fender-type circuit. Probably worth experimenting with lower values - at a guess you want about 2W total for both plates or the reverb might sound a bit harsh because the valve is cutting off some of the waveform.

Great, I'll try some lower values of Carvin R24.

ICBM said:

Keefy said:

There is now plenty of reverb available. There is a noticeable increase in background hum as the reverb level is raised. I will spend some time poking wires with a chopstick to see if I can reduce this with lead dress.

Try changing the orientation of the reverb tank, it's most likely just hum pickup from the return transducer. Make sure the Output end is at the opposite end of the amp from the power transformer first...

See my update post - it was my Variac! Roger that about tank orientation though.

ICBM

I missed the updated post... the level of reverb you'll get from a plate-driven send like that is far lower than with a proper reverb drive transformer, no matter what you do. Increasing the value of C26 as far as possible may help - but bear in mind it must be at least 400V rated.

Keefy

ICBM said:

Keefy said:

I calculated that the two triodes in the 12AT7 driver are only dissipating 0.54W between them (as against an absolute maximum of 5W) so it's not going to be red-plating any time soon!

Ah... I may have miscalculated. The voltage drop across the 47K plate resistor will be much greater than the OT primary in a Fender-type circuit. Probably worth experimenting with lower values - at a guess you want about 2W total for both plates or the reverb might sound a bit harsh because the valve is cutting off some of the waveform.

Did you mean lower values of the driver plate resistor Carvin R25?

At the moment I am finding that whatever I do to the Carvin cathode resistor R24, I always end up with about 0.5W dissipation from the two 12AT7 triodes - reducing R24 increases the quiescent current, but also causes a bigger voltage drop across R25, and these effects cancel out.

I realise the circuit has its limitations but I just want to get the most out of it without cooking anything. There is enough reverb available for regular playing - if I want to go all Dick Dale, i will break out the Princeton reverb!

ICBM said:

I missed the updated post... the level of reverb you'll get from a plate-driven send like that is far lower than with a proper reverb drive transformer, no matter what you do. Increasing the value of C26 as far as possible may help - but bear in mind it must be at least 400V rated.

I may try a higher value of C26 may allow more lo frequencies through so I may give that a go.

ICBM

Keefy said:

Did you mean lower values of the driver plate resistor Carvin R25?

Sorry, confused you! No, lower values for R24 - but still higher than the original 220R. Reducing the plate resistors would actually reduce the voltage swing available to the reverb send.

Keefy said:

At the moment I am finding that whatever I do to the Carvin cathode resistor R24, I always end up with about 0.5W dissipation from the two 12AT7 triodes - reducing R24 increases the quiescent current, but also causes a bigger voltage drop across R25, and these effects cancel out.

Interesting, I would have expected more power with the lower R24 values.

You may need an oscilloscope to find the optimum value, in that case - you're looking for the greatest signal swing between cutoff and clipping across R25.

Keefy

ICBM said:

Keefy said:

Did you mean lower values of the driver plate resistor Carvin R25?

Sorry, confused you! No, lower values for R24 - but still higher than the original 220R. Reducing the plate resistors would actually reduce the voltage swing available to the reverb send.

Just checking! Yes, that make sense.

ICBM said:

Keefy said:

At the moment I am finding that whatever I do to the Carvin cathode resistor R24, I always end up with about 0.5W dissipation from the two 12AT7 triodes - reducing R24 increases the quiescent current, but also causes a bigger voltage drop across R25, and these effects cancel out.

Interesting, I would have expected more power with the lower R24 values.

You may need an oscilloscope to find the optimum value, in that case - you're looking for the greatest signal swing between cutoff and clipping across R25.

I have had R24 down to 295R (combination of existing and piggy-backed resistors) and that gave me:

B+2 331V
12AT7 plates commoned at 125.8V
12AT7 cathodes commoned at 1.25V

By my calculation this gives a combined quiescent current of 4.14mA and dissipation of 0.515W.

Unfortunately I don't have an oscilloscope, except for a DIY thing off eBay that was a complete waste of £17!

Keefy

I've been thinking some more about this reverb driver. It works, but I want to make sure it works the best it can within the limitations of the type of circuit.

In the scribbles below, (1) is the original Carvin design using half of a 12AX7. Plate voltage is spec'd as 210V, which would make the plate current about 2.7mA and dissipation just over 0.5W.

(2) shows a second triode added in parallel, with resistors.

(3) is electronically equivalent to (2), so if I had just used both halves of a 12AX7, it seems to me that R24 and R25 should have halved in value to keep both triodes properly biased.

Obviously a 12AT7 has different characteristics, but I feel like I'm heading in the right direction here. It's a bit late to get the soldering iron out now, so in the meantime I have hot-glued any flapping leads, and have dug out my eyelet punch with a view to making a leatherette bag for the reverb tank.

Keefy

Well that was interesting! I tried lower values of R25, and then various values of R24, by attaching other resistors across the installed ones using croc clips. None made the reverb sound any different so I have left it as-is and fitted the chassis (now with labels for all the controls and fuses) back into the cabinet.

I’ve been busy with other stuff the rest of the day but this evening I hope to make and install the reverb tank bag. Then I’ll call it finished.

Pics tomorrow!

Keefy

Here's the reverb tank bag. The tank slides in from the left and then a flap of leatherette tucks underneath it. I secured it to the floor of the combo with screws through the bottom left and top right eyelets.



Here's the finished article. At low volumes there is enough hum that it bugs me, but once you crank it, all such considerations evaporate. I'm l;ooking forward to trying this at an open mic next week, as long as that doesn't turn out to be illegal.

ICBM

lol at 'ICBM 30R'  

Keefy

ICBM said:

lol at 'ICBM 30R'  

I wondered if you'd notice that! I chose the name in recognition of your invaluable help, and because it's a kind of rocket!

Thanks again 

Keefy

Er... you know that alleged 12AU7 I've been using? Turns out it was the 12AT7 (marked ECC81 - I really should learn those equivalent codes) that I'd put by for another build!

EDIT: see post below!

So I'll be trying the circuit with a paralleled 12AX7 as the driver, or maybe one of the 5963's that I've got lurking in the bottom of my valves box...

Keefy

Having read back through this thread, I realise the driver should have been either 2 triodes of a 12AX7 in parallel, or a 12AT7 triode. The paralleled 5963 made the whole circuit noisy so I have put in a 12AX7, which works fine. I'm going to put R24 back to 220R at some point and re-check the voltages.

Keefy

Another incremental win tonight.

I was moving the MOD reverb tank around to see whether I could find a less hum-prone position for it, but I couldn't. I already had it oriented with the output transducer further from the PT. 

I tried an old Folded Line tank instead, and bingo, no hum!

Keefy

By way of an update, I recently used this amp on a couple of rehearsals, and it sounds brilliant - loud, punchy, fat and bright, all in a small lightweight package.

Keefy

I recently re-worked the valve driver circuit to that of a Magnatone, which employs both triodes of a 12AU7 to drive a 4F… tank. There’s definitely more reverb level available so that’s another win.

PeteC

Hi Keefy
How are you finding the Rocket style amp ? 

I have been revisiting my Express as I wanted to build a new cab for it and I have tamed the oscillation I was getting after reading a hell of a lot of the old posts across at Ampgarage.  

I am thinking of building a Rocket (or even a VVR Express) this winter and wondered how you like it, and what kind of character it has? 

The Express is a bloody loud amp - even with a pair of 6V6s in it ( at 400v plates ) It can go from sparkly clean to roaring overdrive from the guitar volume, but its not something I would dare take to one of my usual club/pub gigs tbh !!!!   I'm looking for somewhere here in North Yorkshire where I can crank my amps to properly road test them. 

cheers
PeteC 

ICBM

PeteC said:

I'm looking for somewhere here in North Yorkshire where I can crank my amps to properly road test them.

Isn’t there a disused RAF base somewhere up there? I seem to remember they tested the recreation of Guy Fawkes’ planned gunpowder explosion on it - should be miles from anywhere and suitable for very loud noises .

PeteC

There’s one about 2 miles from me at Dishforth - but they’ve since made it home to a logistics regiment so that’s a no-go ( literally ! ) 
I might see if anyone on the board has a rehearsal space in N Yorks that would do.