Guyatone (toyo sound) GA-440 reverb tank in need of repair...

[kevtronic]

I have an old Guyatone (tokyo sound) GA-440 and the reverb tank that is broken. I have attached some pictures. On one side the little metal ribbon is broke off, an easy fix with some solder. On the other side the piece that attaches to the spring is broke of at a fold. I'm thinking it can be soldered or maybe use a small piece of copper shielding tape? There are flat bars on both sides that attach to the thin ribbons and then goes through the piece attached to the the spring. Should there be one flat bar per side or two per side? I ask because on my tank there are two per side, but it almost looks like both have been broke in half. If there is supposed to be one per side can they be repaired some how?

With that said, has anyone out had any experience with these tanks that may be able to give me some repair tips, tech tips on testing it, or maybe a source to buy a direct replacement? If it can't be repaired and I can't find a direct replacement any thoughts/tips on an alternate tank replacement? Maybe one that's not connected to the chassis and sits on the bottom of the cab like most amps have them? Thanks for reading and I look forward to reading your responses...

[nickb]

Can you post the schematic?-We should be able to suggest a suitable replacement from looking at it.

EDIT: Uh, I just saw Bill's post about these being ceramic in your other thread. Still, I've seen conventional electro-magnetic one driver directly by the tube - I think it was a Carvin Bell-Air. it used a 9GB2C1B, IIRC. Anyway, post the schemo and I'll have a clearer picture.

[kevtronic]

thanks for the info. This shematic is for the GA-430, but from tracing it against the the GA-440 the reverb, tremolo, tubes, etc are wired up the same: GA-430-trimmed.pdf

[nickb]

Thanks.

The thing that stands out to me is the lack of a reverb recovery stage. So, the piezo unit must have a much bigger output ('lower insertion loss') that the EM ones that I'm familiar with.

I think to make a EM one work you would need to add the recovery stage. The drive current would be OK if you went with as high an impedance input coil as you can get. That would be a "G" with something like a 10k input impedance. The drive stage has a source impedance around 50K ohms which will give you a fairly constant current drive which is desirable.

[kevtronic]

I was under the impression that the second half of the 12ax7 was being used as the recovery stage for this amp? If not, your saying I basically need to add another triode between the output of the reverb and the second half of the existing 12ax7?

This 'ceramic transducer' reverb tank is a first for me, so, I'm on a learning curve. I've repaired quite a few amps, but as far as reverb repairs go it's been tubes/tanks/corrosion/etc thus far...

If I do the conversion it seems like replacing the 6av6 with a 9 pin preamp socket (12at7/12ax7) for the driver would make sense and if I can't use one half for the driver and the other half for recovery then a second 9 pin socket (12at7/12ax7) can be added. Am I heading in the right direction or am I driving off a cliff here? I'm more than willing to learn, buy the parts, and do the work. I bought this amp in broken condition for 50 bucks as a project.

It's a cool old amp that deserves a second chance and a good home
And, to be honest it's a great sounding amp for my tele

[nickb]

I think you are on the right lines. Carvin used a single 12AX7 for the job - look here

I cam across this recently as one needed a replacement reverb tank it was a 9GB2C1B. I didn't have one in stock so used a 4FB3C1B - longer decay with 2 springs and about 1/4 of the input impedance and it worked fine. About half the drive but it was still plenty.

[kevtronic]

I think you are on the right lines. Carvin used a single 12AX7 for the job - look here

I came across this recently as one needed a replacement reverb tank it was a 9GB2C1B. I didn't have one in stock so used a 4FB3C1B - longer decay with 2 springs and about 1/4 of the input impedance and it worked fine. About half the drive but it was still plenty.

I was taking a look at the schematic (guyatone) and had a thought. Why not replace the 6av6 with another 6bm8? Feed from the 1st half of the 12ax7 preamp into the pentode half of the 6bm8 for the reverb driver and use the triode half of the 6bm8 for the reverb recovery and feed it into the second half of the 12ax7 preamp? I could be wrong but this should provide plenty of drive and recovery for a more readily available EM type reverb tank? I was thinking of just dropping in an el84 for a driver, but that wouldn't do much for the recovery (meaning adding a socket), where as the 6bm8 would only require replacing a socket (socket wise/tube wise, not considering bias, etc, which should only require exchanging component/values).

Thanks for taking the time to go thru this with me.
If nothing this project is going to make me take a more in depth look at reverb tanks...

[nickb]

That's a good idea. It will certainly give you a lot more drive current and the 12AX7 IS a bit wimpy. A 2mA rms drive current is suggested for a 4FXXXXXX tank but I know they can take a lot more than that as I did some tests a while back.

I'm guessing the B+ comes out around 400V so if (say) you use a 10K 7W plate resistor and run the pentode at 20mA you should get a respectable drive current through the tank. The drive from the first stage to the pentode looks in the ball park to make this work without over driving the pentode. This will give you something like a 7k source impedance for tank which means it will behave a bit like a constant current source - this is good thing. You could improve on things by using an inductor instead of a 10K resistor but that is a rather expensive option.

On a different tack, can I suggest you replace the half wave rectifier with a full wave bridge? It will halve the hum, reduce the power lost in the transformer and lower the ripple current in the reservoir cap. This improvement in supply efficiency will help to supply the extra watts needed by the pentode.

Good luck, have fun and remember to practice safe modding.

[kevtronic]

That's a good idea. It will certainly give you a lot more drive current and the 12AX7 IS a bit wimpy. A 3mA rms drive current is suggested for a 4FXXXXXX tank but I know they can take a lot more than that as I did some tests a while back.

I'm guessing the B+ comes out around 400V so if (say) you use a 10K 7W plate resistor and run the pentode at 20mA you should get a respectable drive current through the tank. The drive from the first stage to the pentode looks in the ball park to make this work without over driving the pentode. This will give you something like a 7k source impedance for tank which means it will behave a bit like a constant current source - this is good thing. You could improve on things by using an inductor instead of a 10K resistor but that is a rather expensive option.

On a different tack, can I suggest you replace the half wave rectifier with a full wave bridge? It will halve the hum, reduce the power lost in the transformer and lower the ripple current in the reservoir cap. This improvement in supply efficiency will help to supply the extra watts needed by the pentode.

Good luck, have fun and remember to practice safe modding.

The amp already uses full wave rectification, which is one thing thing different about the GA-440 amp I have compared to the GA-430 schematic above.

I have two reverb tanks on hand. One is a 2 spring Altronics, but is just too long for the cab at 17" in length. The other is a none branded 2 spring in a 9.25" tank that will fit nicely. The only markings on the tank is '79201 0'. It has a DC resistance reading of 25.7 on the input and 210.5 on the output. Hopefully I can make it work...

I am going to try and keep this thread going, hopefully, resulting in a complete fixed/moded Guyatone GA-440 amplifer, along with the docs/schematics/etc used. I'm sure I'm not the first, nor will be the last, with these issues for this cool old amp...

Thanks again for the help and input...

[nickb]

That sounds alike 1BBXXXX tank. It needs about 3x the drive current as the drive coil has less turns:

[kevtronic]

That does seem to be a close match. Thanks for the link to the pdf. So, it looks like the 9.25" 2 spring tank I have on hand will most likely need about 6.5mA of AC current to drive it. I'll have to do some checking to see how much AC signal the 6bm8's are providing...

[kevtronic]

I am going to attempt to add a 'transformerless' single tube driven reverb to the amp. I'll be replacing the 6va6 tube with a yet determined dual package tube and use a yet determined short reverb tank. That's informative isn't it - lol. Anyways. I'm on a learning curve and doing some reading/research...

[Leo_Gnardo]

"Dual package" as in one triode to drive and another to recover, or paralleling the triodes for more current drive?

First way, I'd recommend 12DW7, one triode acts like a 12AU7, the other 12AX7.

Second way, you could do as Ampeg has in their 60's amps: 6CG7/6FQ7 or a bit smaller 12AU7.

Either way couple plate(s) to tank with a big film cap. I recommend putting in a bleed resistor at the tank end so you don't take the chance of discharging the coupling cap thru the tank drive coil and turning it into an expensive fuse.

[52 Bill]

Are you sure that the original tank is dead and un-repairable?

[kevtronic]

Are you sure that the original tank is dead and un-repairable?

I was hoping to repair it, but I'm pretty sure it's not able to be repaired short of replacement parts. Check out the last picture above. It's pretty hard to see, but on each end of the spring there are two little, what appear to be ceramic flat bars (not sure how else to explain it) that are broke in half. Correct me if I am wrong, but I believe there should be one per side, not two ( due to being broken in half)? One side of the flat bar attaches to the red wire, the other end is attached to the retainer the flat bar is connect to, and the retainer itself connects to the white wire via connection through the main body. I'd love to be able to repair this if anyone knows how...