Beginner project: Heathkit A7 into guitar amp

[J Martin]

I own one of those amps and I just restored one for a friend. Those amps don't need much of any mods. I wouldn't waste time trying to convert it to a Fender, you can buy Fenders off the shelf. My procedure was to convert the single triode (12SK7?) to a dual triode 12SL7 and make the new triode guitar friendly. It sounded really good. I did all the caps as well. That alone increased the volume quite a bit.

I just bought a whole box of 12A6 tubes at a hamfest, so I will be doing some experimenting. My Heath, and my friends's both had 12V6 as the power tube. Later versions, I guess.

[G-hoppa]

Well, I can't get a Fender 5C5 plus hours of soldering fun and learning off the shelf for the little I spent on this Heathkit. I was looking forward to the experience more than anything else.
That said, I am definitely coming around to your point of view -- listening to PRNDL's recordings helped me decide to bring this back to life as an essentially stock A7 with a few slight mods, to the input circuit and the tone. I got the manual for this thing in the mail today-- so cool. Even tells exactly how to disable the RIAA equalizer to get a flat (and higher gain) signal off the 2nd input.

Funny you mention the 12V6s--there's a note by the original owner of the manual that indicates the only change necessary to use 12V6s instead of 12A6s is to swap the 2W 470ohm cathode biasing resistor to an 800ohm, or throw a 330ohm in series with the original. (am I right about that resistor's purpose?)

What do you think the tone/wattage difference would be between the two power tubes?

[loudthud]

I made some changes to the schematic with Windows Paint. Changed a couple of things like rectifier connection to make it GZ34 compatable. This thing will have a heck of a lot of gain. Rearranged the ground per Kevin O'Connor. Added a fuse at the last minute.

[PRNDL]

PRNDL-- what components on the A-7 constitute the RIAA ciruit?

All the RC's after the first gain stage, except for the coupling cap.

I thought the circuit looked more like a 5e3, with the cathodyne phase inverter.

Also realize that the speaker (an EV SRO Alnico) is a big part of that tone.

[PRNDL]

This thing will have a heck of a lot of gain.

That's true ... way too much gain for input 1 and not enough for input 2.

What I did was use only one input and rewire it to switch between the two gain stages, one with half the gain from the 470K/470K resistors.

I also clipped the 200MMF cap on the PI grid, which is there to limit HF oscillation. There wasn't any perceptible difference, but I didn't care to solder it back in.

I may have changed the feedback to the 4 ohm tap to give it a bit more grit, but NFB makes amps have better cleans.

As for matching 12A6's ... back then I didn't know enough to ask for that or check the bias. I may have been lucky, but 12A6's in metal cans are rugged!

12A6's have the same pin out as 6V6/6L6, so you can use any of the standard bias meters to match them.

[G-hoppa]

loudthud-- I see you've routed the filter cap grounds to the cathode grounds of the tubes they feed--what is the rationale behind this approach? Controlling the current path to avoid . . . what exactly? My grasp of grounding concepts is shaky.

[loudthud]

There are really two reasons to separate the grounds. First, the ground of the first filter cap after the rectifier is the noisiest ground in the whole amp. It's the buzzy kind of noise at twice the line frequency. A push pull amplifier stage can reject quite a bit of this but a single ended stage can't. By separating the grounds, we make sure that the buzz doesn't get on the ground side of the down-stream preamp filter caps (relative to the ground at the preamp stages). If there is noise on the ground side of a filter cap, the cap conducts it to the B+ side with almost no attenutaion.

Secondly, there are small signal currents that must flow from B+ to ground (through the filter cap) at each preamp stage. Having several stages on one filter cap means there can be some feedback from one preamp stage back to the one before it. This can cause an oscillation if the gain is high enough. So it's best to keep the path for these currents as short as possible. Sometimes it becomes necessary to add another RC filter to isolate two preamp stages.

If the high gain is too problematic, try substituting a 12SN7 (mu=20) for the 12SL7 (mu=70). Other than that, removing one stage or adding attenuation between stages should control the gain to workable levels.

[G-hoppa]

What I did was use only one input and rewire it to switch between the two gain stages, one with half the gain from the 470K/470K resistors.

PRNDL-- can you break it down a little for me-- what does this look like? You mentioned a push-pull pot for the high/low gain selection. So, would this bypass the first triode in the 12SL7? And the pot's output would still go to the 12SQ7?

[PRNDL]

PRNDL-- can you break it down a little for me-- what does this look like? You mentioned a push-pull pot for the high/low gain selection. So, would this bypass the first triode in the 12SL7? And the pot's output would still go to the 12SQ7?

Only one of those two gain stages are needed, which means there is a choice:

- Matchless Spitfire left the 1/2 a 12AX7 unused
- Marshall wired each side to a pair of inputs - high and low, which many jump in parallel
- 18 Watt Light II wires to a pair of inputs - single and parallel
- Fender Pro Junior is similar to a Spitfire with the extra gain and a divider network
- Some amps switch in an extra gain stage for distortion

There's no room on the front panel to put two input jacks or an additional volume/tone for the distortion channel. It was a bit complicated, but I used a DPST push/pull pot to switch between one of those gain stages. The idea was to add a little bit of boost. Initially I used the existing pot, but the slider switch wasn't that great, so I replaced it.

As you can tell from the sound clips, the amp on 10 gets a decent amount of distortion. You may want to try having both gain stages, which will give tons more gain for metal. The risk is too much gain causing a nasty buzz.

[G-hoppa]

I made some changes to the schematic with Windows Paint.

loudthud-- belated thanks for this. And for the explanation of the grounding scheme. this is invaluable information for me at my basic level of understanding.
PRNDL--thanks to you too-- I will cook up something along the lines of the Spitfire, I think. No need to aim for metal distortion.
Now I'm just waiting for some parts to get started . . .

[CitizenCain]

Say PRNDL, what values did you use to tune the tone stack for guitar? I have an old Heathkit AA-23 that I'm converting for guitar use. The tone stack is contained in small plastic-encased module. Itwould be hard to change any component values in it. If the stock onesa re close enough, I'll just use it as is.

[PRNDL]

what values did you use to tune the tone stack for guitar?

I used the tone stack calculator to match the curves to the existing pots.

These are the values for the Heathkit
Resistors: 100K, 33K, 100K
Caps: .002, .02, 200p, .002
Pots: 1M-A, 500K-A

These are the standard values for a new build
Resistors: 68K, 56K, 220K
Caps: 500p, .0047, 500p, .001
Pots: 500K-A, 500K-A

The main issue I found was designing the tone stack with values of caps that are readily available, so that you don't have to buy 2 to get the right value.

The problem with the James circuit board is you can't tell what the values are inside. The simplest method is to remove it, and use new caps and resistors.

[CitizenCain]

Those are interesting values. I ran them through the tone stack calc, neat curves. Yep, I think just creating a tone stack would be simpler than trying to modify the little circuit card these amps use.

[G-hoppa]

I made some changes to the schematic with Windows Paint. Changed a couple of things like rectifier connection to make it GZ34 compatable. This thing will have a heck of a lot of gain. Rearranged the ground per Kevin O'Connor. Added a fuse at the last minute.

Loudthud-- I notice on this you've changed the B+ for the 12SL7 from 180V to 330V. Is this the mod you made to enable the use of a GZ34? Since I'm going to keep the 5Y3, can I simply use the original 100K resistor there at the last filter cap?

What would such a large voltage difference have done to the gain and tone of the 12SL7?

[G-hoppa]

trial schematic--suggestions?

okay-- here's the schematic I'm going to try-- please take a look and let me know what you think. I've kept Loudthud's grounding scheme, borrowed the tone stack values PRNDL mentioned, and bypassed the first (unneeded) gain stage. Not sure about that first gain stage, though . . .

Let's see if I can figure out how to upload an image.