GA-19RVT build

[mmaguire9]

Hello to all

I thought I would avail myself of some of the profound expertise here on the forum.
Ever since I repaired one of these amps, a 1960 tweed using this schematic:

https://guitarwork.ru/electronic/firm/Amplifier/Gibson/Gibson%20GA-19RVT.jpg

Not the more common “Crestline” schematic in Doug’s library. Definitely this circuit (as there seems to be a lot of confusion about the proper circuit), I wanted to build one.
I’m now thinking of offering a clone of this amp.

I’d like to keep mods to a minimum and build with all NOS tubes (which I have).
I have a few questions for the folks in the know here. I’m confident in building the amp but there are some bumps on a 60 year old circuit and I’d appreciate any comments on my possible workarounds or suggestions I have not considered.

1.   All the bypass caps in this circuit are 20uF (25v?). I have not found any low voltage 20uF anywhere. I usually use the Sprague Atom 25uf/50v caps here.
Would there be any noticeable change in sound? I’m thinking not since most caps are +/- 10% anyway. Even so 5uf is 20%. I’m not anticipating any real changes here but I’d appreciate thoughts.

2.   As far as I can determine, there are No 2M audio taper pots made by anyone as far as I can determine. There are a few 2M pots but all are linear taper, unsuitable for this application. What can I expect if I use 1M audio tapers in those volume controls?
Especially on the normal channel, that would change the input impedance to the grid somewhat. Not sure what to expect. Surely, if someone knows where to get 2MA pots I’m all ears. I don’t believe I can set up any voltage dividers that will still give me full volume sweep. This is probably my biggest concern on the build. Comments or suggestions?

3.   My only option for a reverb transformer is from Mercury Magnetics :-(
Pretty pricey at $68 for a 3 watt reverb transformer. In addition, I would have to get 4F reverb tanks instead of the 4A which isn’t a real issue but now I have to stock tanks that can maybe only be used for this build. The doubled up 12AT7 driver in a BF reverb circuit should be around 22K at the output and the BF transformer usually is specified at 25K-8ohm.
The output of the 6U8A (in place of 7199) I’ll be using is about 5K. Too big a mismatch to use the 125A1A. The unused triode of V3 is available but has an output about 62.5K so again, that’s a big mismatch plus the gain is much higher.
Anyone know how I might use the 125A1A and the 4A tank here? I can’t think of a solution, but if anyone has sussed this out I’d be grateful for the knowledge.

4.   I was planning on using a Hammond 290PAZ for the PT and a Hammond 1760H for the OT. I believe the PT is used in the Marshall 18W. The OT lifted from BFDR should be gtg and allows me to offer different speaker options.
I notice the 6.3 filament circuit calls for the CT to go to the 6V6 cathodes. I’m assuming this is to elevate the filaments and help with hum reduction. Any issues there?
In addition, should unmatched 6V6s be a concern? They are both drawing off the same cathode resistor and will affect the final drop across the 270ohm there. I have not built a self biased amp before but I have worked on a ton of radio gear that uses it so I’m not unfamiliar with the operation but it’s been awhile.

5. Using the 290PAZ at 290vac out through an NOS 5V3GT into a 20uf filter should get me close to the 300v called for the power tubes. I don’t want to go Fender high here but would like to stick close to the  lower voltage and not have to play around with finding a different cathode resistor value if I can help it. How does that look?

6. Finally, any thoughts on adding an NFB loop to the circuit? Would the benefits be worth the modification?

I certainly will appreciate any guidance the group can provide and I will be glad to share pictures of the prototype build as it progresses.
Hope everyone had a nice St. Paddy’s Day

[mmaguire9]

Just an update to above ref: 3

Unless I've misread something this transformer could be used as a reverb driver transformer.
connection to the BLK / YEL leads on the output will give me 5K-8 ohm to the tank.
This should be a good match with the 6U8A triode output and allow use of a 4AB3C1B tank.
I like the idea

[Keppy]

I think your biggest problems using 1M pots will be the tone control rather than the tubes, since the volume and tone controls are interactive. You can generally preserve the response of RC filters by making opposite changes to the Rs and Cs, so if you cut the Rs in half you can double the Cs to get the same response. Based on this idea, if you're halving the value of the volume pots, it seems like you should halve the tone pot as well, and double the tone caps.



You could double the 0.02uF coupling cap as well to be thorough, though it seems like that one is already big enough to pass all audio into a 500k impedance.


You might lose a bit of gain because those pots are effectively in parallel with the V1 plate resistor for audio frequencies. My math says that the plate resistance at audio drops from ~180k to ~150k by halving the pots. That's probably not a noticeable difference.

[mmaguire9]

Hi Keppy

Thanks for the advice. Simple and elegant. I will do that. Caps are easy enough to go in and out if I have to adjust them. Maybe a 0.033 for the V1 coupling cap?
Appreciate the comment 

[AmberB]

My suggestion would be to use at least a 5 watt resistor for the output tubes cathode bias resistor.  The 2 watt specified on the schematic is probably not heavy enough if you play the amp loud often.

[mmaguire9]

Thanks AmberB for the suggestion. It's not a bad idea. Should be plenty of room to mount it and get some air around it. No signal would still need a 2 watt or bigger but when the big peaks come, who knows?

[mmaguire9]

Well, I'm getting deep in this project and I want to thank the folks here for turning me on to JSchem.
I run Linux here but Java is the Linux users best friend.
If I could get something like JSchem for laying out chassis, I'd be all set.
For now, it has enabled me to get this schematic of the modified Falcon done.
I'm going to label it a Black Falcon because I hate working with Tweed and the final will get the black bronco treatment.
I'll be doing this in two stages, first a prototype on a big chassis. I wanted a 17x8 but they are out of stock so I got a 17x10. The luxury of space is good for prototyping and once I know the circuit is good, I'll see about stuffing it in a 17x4x3 to make a combo. I'm using a standup PT for the prototype but I'll need to use a laydown for the combo.
Waiting on parts atm but I have all the tubes NOS and the chassis and PT. Sockets, jacks and switches can come out of the bin.
Once I make some genuine progress on the prototype, I'll post some pictures here.
If nothing else, someone in the future might find it a good guide to building their own.
6EU7s are getting expensive and I'm still seeing 12AX7s available. They are about equal in price atm but who knows what the future will bring? That's kind of why I was looking for an NOS project.
Maybe the combo will end up with 12AX7s in the end.
If anyone knows of a decent replacement for the 7199/6U8A with a 12v filament I'm all ears.
Pinouts aren't an issue with a new build but EPR might be.

If anyone checks this schematic out and notices any mistakes I'd appreciate hearing.
Spring is almost here 

[tubeswell]

Another suggestion for reverb


Why? - The reverb return signal from V2b will be too weak to inject after the V3a, and would be better going to after v1a instead. Also you should have a reverb level pot after V2b to adjust the wet level.


I didn't have time to check all the other bits of your schematic for errors sorry. (I only looked at aspects that looked immediately obvious to me)

[mmaguire9]

Hi Tubeswell
Thank you for taking the time to go over the schematic and for your thoughts.
My thinking was to keep the original circuit as close as possible. I may have made an error in redrawing the circuit from the OP and I'll definitely check it over again. I verified connection by connection the original amp I had here in the shop and was satisfied that it was in fact that original circuit as drawn.
I had no trouble with the reverb but it was quite a bit different sound and operation from say, a BF Fender.
I am curious why you do not think the pentode will have enough drive since it appeared to work pretty well (but as I said differently). I have not done all the hard math on that part of the circuit since it seemed straightforward. I was too concerned with matching up the driver to my tank of choice 
For instance, one quirk is that when you 0 the main volume, you still get sound from the reverb channel (easy enough to see why). You have to 0 both VOL and REVERB to silence the amp.

Your suggestion does look solid to me and if I do find the reverb not adequate it would be easy to implement on the build. It is in the folder with the rest now.
Thanks again for your work on this

[mmaguire9]

btw

Correction on my previous question about 6U8A alternatives

I guess 10 pins out on a nine pin tube might be a showstopper to a 12v triode/pentode

We always think more clearly away from any computer 

[wsscott]

Just wondering on how the Amp build went.  I'm thinking of building this amp also, and I'm interested in your results. 

[chocopower]

hi there.


I dont know about this particular build, but i have an original crestline Ga19rvt power transformer 
and since i live in Spain with 220vac wall supply, i just cant use it for nothing.


If you want it, i can send it to you.

[wsscott]

Thanks, but my buddy who was interested in doing the build with me decided since he already has 6 amps his wife would be very unhappy to see another one taking up space in the house!  So I’ll let someone else take you up on your offer.

[passaloutre]

I’ll take the PT if no one else has claimed it!

[chocopower]

Ok. Let's talk privately.