AO-43 to Gibson Tweed RVT

[vampwizzard]

Hey folks! picked up an AO-43 pretty cheaply and its conversion to a 6V6 amp was intriguing enough to get designing. Ive seen Sluckey's plexi and was inspired.

Ive attached a PDF draft of the circuit. Im thinking something swart influenced. 12DW7 reverb, half tube trem, tweed style controls.

For reverb, ive borrowed heavily from: https://guitar.com/guides/diy-workshop/build-tube-spring-reverb-unit-amplifier/ which seems to be copied from the AST but with added controls. Since there is only a reverb volume control on those amps with a set dwell value and no tone knob those are left out of my circuit.

another inspiration is from the Atonic schematic floating around https://music-electronics-forum.com/showthread.php?t=22699 but it appears that the reverb circuit is incorrect, but it picks up the set point of the dwell (if they're correctly reading the gutshots)

From the AST guts its clear that the Gibson Scout influence was in feature set only. He very clearly uses a fender style input triode and then the cap coupled reverb. For the rest of this circuit I based it on a tweed 6G2 until discovering it wasnt cathode biased and ultimately found the GA18T schem http://djlservice.com/GibsonGA18wPhotos_files/image014.jpg

So im all over the place. The power circuit is just there for a shortcut for actually designing it later. What do you all think of this mess so far? Have I tied the reverb/trem in correctly? I may change the trem circuit to the GA18T specs or leave it as is. Still undecided on the single reverb control. 

Stay safe!

-Bill

[tubenit]

I am not familiar with a capacitor driven reverb, so I am not sure what the values should be.  The 22k grid resistor on return seems odd?  Usually there is a 220k there.    In ARCHIVES, we have a lot of info and schematics on one tube reverbs most using a reverb transformer.  Might check those out & compare.  The example I've attached is in a working amp I have and I'm super pleased with the reverb using a 12DW7.

As far as insertion points, I think what you've indicated will work but would be impacted by how your volume pot is set.

With 6V6's,  I'd recommend .047 instead of .1 post phase invertor coupling caps. 

With respect, Tubenit

[vampwizzard]

Thanks for giving it a look Tubenit, I was hoping you’d comment on it.

Allegedly that reverb circuit is a working circuit from that guitar.com article. I’ll try and cross reference it to the ampeg circuit its allegedly based on.

I’ll likely try .047 and .022 coupling caps to find a bass response I like.. 0.1 in tweeds seems to be problematic based on my 5e3 build.

I’ll wrap up some revisions tonight.

[PRR]

> 22k grid resistor on return seems odd?  Usually there is a 220k

It makes a small difference. The return coil impedance rises to dozen of Kohm at the top of the audio band. For Most Treble Signal you use 220k. If that is too bright, try 22k to load the top end.

[vampwizzard]

Hey folks. Ive attached revision A. Looking at all the variations on the tweed circuits its amazing to see what folks designed back in the day. For instance, the GA-18T put NFB in on the second preamp triode and dropped the cathode cap value super low so it didnt soak it up. Looks very similar to the situation Rob Robinette discusses in his 5E3 NFB mod.

Dropped the coupling cap values to 0.02 from the initial 0.1.. i did this on my 5E3 build to reduce bass. Shouldnt make the same mistake twice.

Ive copied Sluckey's filtering scheme from his 6V6 Plexi conversion because we are using the same iron. I intend to use a 5Y3 and ive made my first node 40mf.. is that still safe with a modern tube or should I go a flat 20 across the board? Also using a choke because it came with the amp already. As far as node sharing, I mimicked what the GA-17RVT Scout did for triodes per node.

Does this seem like a viable project to start laying out? Have I made any egregious errors?

[echuta13]

R16 value doesn't look right.   I'd be curious to see what wattage values you are using in some areas as well.  R22 should probably be a bit beefier.  Have you thought about led biasing the trem triode?  I think the .02 on the coupling cap is a good move.  :0)

[PRR]

> 40mf.. is that still safe with a modern tube or should I go a flat 20 across the board? Also using a choke

If you have a choke, why do you want a monster first cap?

[vampwizzard]

R16 value doesn't look right.   I'd be curious to see what wattage values you are using in some areas as well.  R22 should probably be a bit beefier.  Have you thought about led biasing the trem triode?  I think the .02 on the coupling cap is a good move.  :0)

R16 is definitely wrong. Should be the standard 250.

R22 is taken directly off the guitar.com article in the OP.

PRR: fair point. Those values and the triode per node were the two main questions I had. 20-20-20-20 is adequate? The main reason I’d opt for the 40 is of how cheap the JJ Van is.

[echuta13]

R22 is taken directly off the guitar.com article in the OP.

That's fine as far as the value goes, but I would spec a 3-5 watt resistor there.

[vampwizzard]

It took me a long time to get DIYLC working correctly on the mac. Its also been a LONG TIME since ive done a component level layout vs schematics. Point to point was a little easier than this. I fall apart where the reverb and trem circuits need to insert. Any advice would be greatly appreciated. Do you minimize number of turrets? Minimize jumpers? Make sure components stay upright or horizontal and avoid the angles?

[vampwizzard]

A different layout. I like this one better. any opinions welcome.

[vampwizzard]

I've had some bad run ins with board layouts. Hoping to do a tweed/blues deluxe/blues jr style chassis layout with controls on top and the tubes and controls went smoothly. Made a template for the PT and dropped an eyelet board in. The spacing is out of control for 29 eyelets. Whats a good remedy for this? Terminal strips? The donor amp board is there for reference.. 20ish terminals in half the length.

[jojokeo]

I've had some bad run ins with board layouts. Hoping to do a tweed/blues deluxe/blues jr style chassis layout with controls on top and the tubes and controls went smoothly. Made a template for the PT and dropped an eyelet board in. The spacing is out of control for 29 eyelets. Whats a good remedy for this? Terminal strips? The donor amp board is there for reference.. 20ish terminals in half the length.

There's nothing like experience to know what you can pull off in the required amount of space available. It is a specialty for some of us. The best advice is to get out your pencil, paper, and a good eraser along with stencils to create actual size drawings for all of your parts you plan to use. This could include actually getting your parts placed in front of you to know what their actual sizes are that you'll be using. Space your eyelets/turrets/tag strips accordingly. Try all methods and combinations of methods to achieve what it is that you want to do. Always plan ahead and there will be no surprises during your actual build.


Something you mentioned also is having problems with board layouts? It is highly suggested that if you use anything by someone else - you follow every single connection from the original schematic and see that the layout is done correctly, and didn't leave certain things out? If so, you'll know exactly what it is you need to correct ahead of time? I rarely ever follow or create something made by someone else unless I follow this advice myself due to NOT doing this a long long time ago on an effects pedal back in the 80's. It's never happened since. A side benefit is that you will get real good at reading schematics and creating layouts of your own, often improving on the originals. Here's an old AO-43 build from about 6 years ago for inspiration.

[vampwizzard]

jojokeo - Thanks for the advice and what youve said makes sense.

Ive done other builds. My first one I reused the hammond board (like the one pictured) for a 5E3 deluxe plus a trem circuit. The original board I went to use didnt work out as intended because of the width of the custom eyelet boards. Second one was a matchless spitfire and I did true PTP. Again, no issues there. The third one was a JTM45 and I was didn't stray terribly far from the original layout and got it to fit in with some nylon standoffs and fun wire runs.

So my problem isn't so much laying out the boards necessarily, its laying it out in the freeform vs laying it out on known grid. I figured the three rows of eyelets would force a design (true) and im happy with what I came up with in DIYLC but how can I translate the 29ish eyelets to a smaller spacing to make it fit widthwise in a chassis? Do I design a custom turret board with the expectation to rip cut an inch off? Or should I mimic the matchless/badcat layouts and run them off turrets tags?

[jojokeo]

If I'm hearing (understanding) correctly - improvise... looking at the last board layout you have a couple of power filter string resistors to the far left. Mount these directly to the cap cans or use a tag strip eliminating these from the board. You have a gap between your cathode bias power tube resistor & cap. There's another gap to the bias resistor/cap next this this. Move it over also you don't need this bridging all the way.

You can likely shave off maybe 11/2" right there? Okay I've crudely moved your parts on the left side to the right and numbered the moves to give you an idea...This drops your width from 29 eyelets down to around 23. There's more you can do on this board earlier in the circuit also...

[vampwizzard]

Jojokeo: I added a very large alcoholic drink and some good tunes in the background to your suggestion of deep thinking. I believe if I split the last four eyelet columns off the board I can break that board out as a B+ voltage board. All of the connections are going to a cap can, tube or ground.

the width would go from 14.5" to 12" and would then easily fit. ive got more than enough space to mount that power board.

Again, very appreciative of the thought power and guidance.

[vampwizzard]

Chassis partially drilled. This is the blank aluminum from mojotone. 2.5" tall panels for controls and tubes (tweed style layout). In librecad drew up a rectangle with a center line at 1.25" then drew the component circles and spaced accordingly. Taped them to the chassis aligning the center lines with a marked center line on the chassis. Worked out pretty well for me. Used a drill press to make the pilot holes then ran it out with unibits. That worked pretty well. Made a couple of errors from not using a drill press in a long time but no worse for wear. Probably going to invest in an IEC punch.

Transformers and cap can are the major items left for the interior.. laid them out with a template hopefully my jig saw and dremel skills are up to the task.

More updates to come.

[sluckey]

 

[vampwizzard]

Sluckey, or anyone familiar with AO43 pt's:

If I wanted too axe the choke and replace with a resistor are there any suggested values to try? How do you ball park them in your designs?

[sluckey]

The choke on the AO-43 chassis is NOT part of the power supply circuit. It's in the reverb driver circuit.

[vampwizzard]

well now im glad i asked! that couldve been bad.

Since im not really looking for a voltage drop between those nodes should I look at a small-ish resistor? 10K? or is it a hunt for an appropriate value?

[vampwizzard]

well now im glad i asked! that couldve been bad.

Since im not really looking for a voltage drop between those nodes should I look at a small-ish resistor? 10K? or is it a hunt for an appropriate value?

It appears to be a hunt. https://diyaudioprojects.com/Technical/Tube-Power-Supplies/#First_Calculations

Into the maths. *blows dust off college degree, it smells of miller high life and that liberal arts honors program*

[sluckey]

Since you do not want a voltage drop I suggest you use the choke. I used the AO-43 choke in my Plexi6V6 and Smoky projects. That choke is 14H and has a DCR of 487Ω. Only had a 3V drop between the plate node and screen node on my Plexi6V6.

[vampwizzard]

went down a rabbit hole of fender tweed circuits. it appears that theres very little tonal difference between the choked and resistored tweed circuits. (5E3 vs 5E4 with 6V6's). Going to give it a go with the choke per your power supply on the plexi and if it sucks ill change it up.

[jojokeo]

I've used the choke many times with only a single issue that was traced more to the choke's condition & issues and that the others that have all worked well since. Use logical thinking... do you know what the specs of the '43's are??? Hint: it powered much more than what's only on that chassis-


You don't use a choke or resistor to give you a tonal difference, what are you thinking?

[vampwizzard]

jojokeo: a major component of the 5E3 is the voltage sag generated by the combination of slightly underrated PT/Power supply in combination with the high requirements to amplify so much bass. This does not appear to be a major requirement for all tweed amps though. Chokes are obviously much more efficient in their filtering task.

I think that folks took the tonal differences between say the brownface deluxe and the blackface deluxe and attributed it to the choke in the blackface when thats only a small bit.

Im after a tweed sound. Thats why I was investigating, and after finding a video comparing the 5E3 and 5E4a Ive decided the choke is fine.

[dude]

That choke dropped 2 volts, like Sluckley mentioned, I used it for filtering purposes, not tone. I used Sluckley's and Hoffman's 6V6 Plexi schematic and just changed the layout to fit. Everything worked out great. The biggest issue is placing the board in the right place, I used 2 1/8" width but at first placed it too close to the pots, couldn't get them out without lifting the board. Use short standoff's, nothing taller than 1/4", leave just enough room to get pots out. The other task was expanding the socket holes. I cut out a rectangle piece and added a plate, if you just drill them bigger the 6V6's will be too close or probably won't fit, everything worked out great, just lay everything out like JoJo mentioned before cutting things. If I had to do things over I might have use a single can in the already hole near the pre-amp but I had a lot of double can's in my parts bin. As you can see things are very tight but well worth the effort, the amp sounds great, I added a LaMar MV to go with the pre-MV, mixing the two volumes and masters is awesome. Your work looks very good, hope I helped some, good luck and if you're thinking of another AO43 chassis built, email me, I have 8 more.   


BTW, the tubes in there are 5881's 

[vampwizzard]

Nice build man! I went with a blank chassis to learn how to do some metal fab. Its been a learning process.

For some reason im getting an error trying to upload a picture so after i reboot or get further along ill post progress pics. Im down to mounting eyelet boards/pots and finishing the wiring.

[dude]

Ha, wrong pic of guts, no LaMar MV in this one

[jojokeo]

jojokeo: a major component of the 5E3 is the voltage sag generated by the combination of slightly underrated PT/Power supply in combination with the high requirements to amplify so much bass. This does not appear to be a major requirement for all tweed amps though. Chokes are obviously much more efficient in their filtering task.

I think that folks took the tonal differences between say the brownface deluxe and the blackface deluxe and attributed it to the choke in the blackface when thats only a small bit.

Im after a tweed sound. Thats why I was investigating, and after finding a video comparing the 5E3 and 5E4a Ive decided the choke is fine.

Thanks so much for educating me on what a Tweed amplifier is all about, I didn't even know their was such an amplifier? 

[vampwizzard]

Jojokeo, im documenting the build and my design decisions for myself and others who come later. I never knew about the 6v6 choked super and 5E3 threads with builds using chokes say that it changes the overall character of the amp. We don't know each other well enough to know your experience, but my answer to your question was intended to be a genuine exploration of what makes a tweed amp a tweed amp and not an attack on your experience of amp building.

[vampwizzard]

I found an error in the trem portion of the schematic. I ripped it from a gibson hawk and changed a fixed resistor to a pot, but didnt depict it correctly.

[vampwizzard]

Also need to wire the speed in the same way. Also removing the 1M resistor thats hitting the middle leg of the depth pot. The circuit was lifted from the Hawk schematic which only had one trem control. The Gibson 18T is a better example of how to wire this up.

[PRR]

> depth pot needs to be wired as such.

Why? What is wrong the way you had it?

[vampwizzard]

I left the wiper open, so it’s just a non-variable resistor.

[vampwizzard]

Tested the amp out, B+ unloaded was fine in the low 400s with a 5U4. Loaded the rest of the tubes in to start taking voltages/biasing and we've got a motorboat. Realized I stuck my spare 5Y3 in another build thats with a buddy out of town. Only had a 5U4 to test. Going to get some other rectifier tubes (GZ 34 and 5Y3) to see if its just sucking out on that 40uf node and may rewire the entire power circuit. Other problems include a 120V pilot light that sucks all voltage when wired in series from the fuse to my power switch and doesn't light in parallel.

Overall this is a F-UGLY build. So much so that Im tempted to gut it and start over on a smaller, more tightly spaced board. Definitely gutting the power section to solve this motor boat.

[PRR]

Motorboating is not usually rectifier.

[vampwizzard]

Motorboating is not usually rectifier.

PRR: filter caps and ground issues right? What else should I be looking out for?

[shooter]

fwiw;
I went duckin on "motorboating" n found a thread where swapping OT primaries "fixed it"  ya get what ya pay for 

[vampwizzard]

fwiw;
I went duckin on "motorboating" n found a thread where swapping OT primaries "fixed it"  ya get what ya pay for 

Shooter, tried this out and didnt solve the problem. I had this issue on the bassman build last year though its a great tip.

[echuta13]

That "fugly" part of the build could be the part of the problem (wire routing issue?)  Site unseen it's hard to say.  You might want to chopstick some of your leads and move them around to see if that helps or exacerbates your problems.   

Sounds like a good time for beer worse case! 



 

[vampwizzard]

That "fugly" part of the build could be the part of the problem (wire routing issue?)  Site unseen it's hard to say.  You might want to chopstick some of your leads and move them around to see if that helps or exacerbates your problems.   

Sounds like a good time for beer worse case! 

I said Eff it. Great weekend for some fly fishing. Headed to Doug's neck of the woods. Brought back some NC beer for such an occasion lol.

That being said, its not in the preamp sections, still present after yanking those tubes. My bets are on a bad rec tube, a defective/poorly wired cap can, or a bad OT, in that order.

[jojokeo]

I say "no" to all that you've mentioned for the Evinrude outboard motor, nor will changing the gas, oil, or propeller resolve it. It's one thing to be parasitic surfing and another to be motorboat wakeboarding. Without good detailed pictures of what you're riding it's impossible to keep playing the guessing game and wasting people's time. I see a set building outside...later shore break boy.

[vampwizzard]

Ill bring an oar and try and keep it between the navigational beacons. ill post pictures when ive rebuilt my power section. Waiting on parts.

[vampwizzard]

Alright, ive changed up the power supply a little bit. Coming off the rectifier B+ im running a 250 ohm 10W resistor to node A, 40uf node on the cap can. Resistors between the remaining nodes are 10K 3 watt. Motorboating still present.

Voltages at B+ nodes:
A- 345
B- 339
C- 266
D- 204

All tubes light up, all ground points seem ok. Is it safe to measure voltages on the rest of the tubes with the loud oscillation going on?

[vampwizzard]

Heres some pictures

[sluckey]

Put an additional 20-40 filter cap directly on the rectifier pin 8, such that there will be a filter cap on each end of that 250Ω resistor.

[vampwizzard]

any particular reason for that placement Sluckey? I wanted to drop the voltage before it entered the filtering. sort of like the lil tiger http://solorb.com/elect/musiccirc/liltiger1/

[vampwizzard]

Also: voltages with a JJ's GZ34S.

Sluckey: ive only 16ufs here at the house. Should I parallel them and insert to the nodes or will 16uf give us an idea of whats going on?

[sluckey]

Try a single 16.