Military projector amp - last edited 9/26/22 - layout help needed

[JMPGuitars]

What value resistor would be better there to get the screen voltages right?

Would the removal of the circuit to the left of the R228/R240 affect both the plate/screen issue as well as overall tube bias? That was the 6aq5 oscillator section for film projector purposes that isn't needed for guitar. It is fully removed so the transformer and circuit have one less tube.

You could use Ohm's Law to calculate the resistance you want. I would still use a similar method that you see in the 18W drawings with 1K resistors to the screens after the node A to B dropping/decoupling resistor.

Thanks,
Josh

[yello]

What value resistor would be better there to get the screen voltages right?

Would the removal of the circuit to the left of the R228/R240 affect both the plate/screen issue as well as overall tube bias? That was the 6aq5 oscillator section for film projector purposes that isn't needed for guitar. It is fully removed so the transformer and circuit have one less tube.

You could use Ohm's Law to calculate the resistance you want. I would still use a similar method that you see in the 18W drawings with 1K resistors to the screens after the node A to B dropping/decoupling resistor.

Thanks,
Josh

I've never used Ohm's Law so would need to figure out how to do those maths...and I don't know what resistance I would want as an end.

I have some 3w 470ohm and 5w 1k ohm resistors on hand that I had purchased for screen stopper use but hadn't implemented yet. I already put 1.5k 1/2w control grid resistors for each 6aq5.

I'm still wondering about the overall bias being too hot. Is that related to removing the unused 6aq5 so now more current is going to the other tubes?

[yello]

I tell you, it's not making any sense .. the screen voltages being higher than the plates. There must be something wrong, like I said earlier.

Thinking back, I have a nagging suspicion there is a wiring issue. The only solder joint that looked non original in the wiring of the amp when I first opened it, was at the first power tube 6aq5 pin 6, the screen. The wiring deviated slightly from the other amp I had in how it connected between the turret board component and tube socket and Cap, but I thought logically it was the same thing. I'll take a second look today and rewire it to match my other one, then test voltages.

The screen stopper might still be called for, but at least looking at and correcting the potential issue first would help!

[Bieworm]

I would try this:
Change the 350 ohm resistor between C216 B and C to 1k5 5W
Add a 1k 3W resistor between C216 C and each screen
Change the cathode resistor to 330 ohm 5W

Deleting the futile current draw of the unused tube doesn't affect anything IMHO, except a little more gain in the preamp.. maybe

[yello]

I would try this:
Change the 350 ohm resistor between C216 B and C to 1k5 5W
Add a 1k 3W resistor between C216 C and each screen
Change the cathode resistor to 330 ohm 5W

Deleting the futile current draw of the unused tube doesn't affect anything IMHO, except a little more gain in the preamp.. maybe

I checked this current amp against another in my shop, same exact circuit. Same mods (short of preamp bypass caps). The wiring is correct, there was not an error as I wondered, there are two variations of how wiring gets from C216c to the power tubes: Cap wired to power tube 2 screen, then wire to power tube 1 screen OR cap wired to power tube 1 screen, then wire to power tube 2 screen. Either case has the wire from R240 (350r) going to power tube 1 screen.

Also, I swapped the caps, tubes, and fuse from current amp to the other one, tested voltages, and I get the exact same scenario - ~330 on screens and ~320 on plates. That must just be how these amps are, though maybe not ideal as Josh pointed out. They certainly lasted through decades of use by the military, and I have several friends playing them without issue both in stock and modded formats.

Is it R240 that is between C216B and C216C? That what it looks like. I can change that to 1k5 5w, will just need to order that part.

When you say to add a 1k 3w resistor between C216C and EACH screen, do you mean 1 resistor and connect pins, or separate resistors for each screen and therefore separate wires from the cap to each screen? I'm trying to visualize it, as my 18w diagram for my ef86 build just uses 1 resistor from cap, then wires screens together. I don't fully understand the concept thus my confusion - I'm trying to increase resistance between C216C and the screens right?

Then it looks like if the above is correct for R240 swapped to 1k5, then that leaves R228 changed to 330r right? Or do you mean change R241, the 220r 8w resistor to a 330r 5w?

If you mean R241, I'll have to think through how I implement that and where I put it. R241 is a long ohmite that doubles as a standoff to hold the turret board in place. I could leave it in place for its mechanical function and add the cathode resistor elsewhere, like down with the 2200 uf cap in the new terminal I put in. Alternatively could I put another resistor in series with it to get it up to 330r or close? I have about 50 10w 50r laying around from a Hammond organ. Probably easier to do option A.

[JMPGuitars]

For the 1K screen resistors, the method in your EF86 is fine, or you could add a second 1K resistor and go directly to the 2nd power tube without the jumper wire you see in your EF86 layout (both 1K resistors start at the same point, and the ends go 1 to each tube). The Tremolo TMB and Superlite TMB docs show this idea, however they're on the turret board in those drawings.

[Bieworm]

Ok, here's a theory..
Since your plates voltages are lower than your screen voltages, even though you have 2x 350R dropping resistors between plate and screen nodes, which seems unlogic... maybe due to the very hot bias on the output tubes they are pulling high idle current, causing a big voltage drop across the primary winding of the OT.
To check if that is happening you could compare the voltages at C216 A (mark Y) with the voltages at the plates (mark X). X should be a few volts lower than Y due to the resistance in the OT primary winding
Though I strongly advise you to put in at least some 1k 3W screen resistors (for safety reasons)... this might not be the culprit. I would clip in at least a 50R 5W resistor in series with the 220R Ohmite to drop that cathode voltage. Maybe do that first and see what it does to the balance between the screens and plates.
For the lifespan of your tubes and in case of failure the damaging of the transformers it's important the screen voltage sits lower than the plates. This is common sense.
And... the 50R resistor is merely a guess, so prepare for tweaking that value.
At this moment your tubes idle at 108%. That should drop to somewhere around 75-85%
Suppose that when you add the 50R resistor and you've got 270R cathode resistance. Just thinking out loud.. the plate voltage rises to 330V, the cathode voltage drops to 17V .. so in theory the tubes would idle at 77.5%, which is fine. I don't know if those voltages would alter that dramatically, that's hard to predict.. but you get the idea.
The bias calculations can be done at : https://robrobinette.com/Tube_Bias_Calculator.htm

Dayn screen and plate voltages.png

[JMPGuitars]

I would first probably consider using a voltage dropping Zener diode to reduce your B+ before doing anything besides adding the 1K screen resistor(s).

[yello]

To check if that is happening you could compare the voltages at C216 A (mark Y) with the voltages at the plates (mark X).

I will check that out! I already unsoldered the screen wires, so I will likely add the screen resistors first, then measure. Unless I need to measure before installing them

For the 1K screen resistors, the method in your EF86 is fine, or you could add a second 1K resistor and go directly to the 2nd power tube without the jumper wire you see in your EF86 layout (both 1K resistors start at the same point, and the ends go 1 to each tube). The Tremolo TMB and Superlite TMB docs show this idea, however they're on the turret board in those drawings.

I'll take a look and implement whichever seems to fit best in the tight space.

I would first probably consider using a voltage dropping Zener diode to reduce your B+ before doing anything besides adding the 1K screen resistor(s).

I've read your posting about lowering the B+ here:

viewtopic.php?f=4&t=25724

How much voltage should I drop? (Which zener diode size to buy, 5w)? Looks like all the places I checked they are out of stock and backordered unfortunately.

[JMPGuitars]

How much voltage should I drop? (Which zener diode size to buy, 5w)? Looks like all the places I checked they are out of stock and backordered unfortunately.

Try one of these: https://www.mouser.com/ProductDetail/on ... EMbw%3D%3D

Get more than one.

I typically drop about 30V to compensate for the wall voltage (and SS rectification), probably a good starting point...especially since it's the only one available.

[yello]

I am struggling to find the best way to implement the screen resistor/s. I like the idea of putting individual resistors for each screen coming from C216C. I'm unsure about how R240 (360r 5w) relates to the screens and want to get it right.

The following drawing shows the stock wiring, and two options, first with one 1k 5w screen resistor, the second with individual 1k 5w screen resistors. Does either work? Is option 2 problematic in regards to R240 creating a little wiring loop? I'm still learning, so a "why" along with direction is always appreciated!



Related question - I've been reading a lot about screen resistors and see that 1k is very common with el84 amps, and that 470r is common with 6V6 amps. 6aq5's are like small bottle 6V6's so wondering if 470r is appropriate? I'm totally open-minded to learn which is better and why.

This is one thing I read, from Rob Robinette:

Since screen current affects the tube's operating point along the transfer curve and changes the shape and slope of the tube's transfer curve it has more effect on power tube overdrive tone than grid current. When tuning an amp's overdrive tone close attention should be paid to the screen grid resistor and screen-to-cathode cap values. Taking the time to experiment with different component values can pay great dividends and help differentiate the voice of your amp from the rest of the pack.

Personally I like as much screen voltage fluctuation as safely possible because it changes the shape and slope of the transfer curve which generates tons of harmonic and intermodulation distortion so I use the smallest ohm value resistors that can safely limit overdrive screen current. My suggested screen stopper resistor values are (one screen stopper per tube) :

6V6: 470 ohm (3 watt or higher)
6L6, 5881: 470 ohm 5 watt
EL84 (true pentode) 1k 3 watt
EL34 (true pentode) 1k 5 watt

These values have stood the test of time for tone and reliability.

[Bieworm]

If you look at other 18W schematics , such as Ceriatone, then you'll notice they use a single 100R resistor from the filter cap to split into 2 wires to the screens after the resistor.
That's fine too I guess.
I put 1k on my 6L6 amps too, so I can put in EL34 when I feel like it. I don't think there's a lot of difference between 470R and 1k for the screens. Sole say it muddies up the tone when you put 1K on a 6V6 of 6L6... but I challenge those people in a blind test.

[yello]

If you look at other 18W schematics , such as Ceriatone, then you'll notice they use a single 100R resistor from the filter cap to split into 2 wires to the screens after the resistor.
That's fine too I guess.
I put 1k on my 6L6 amps too, so I can put in EL34 when I feel like it. I don't think there's a lot of difference between 470R and 1k for the screens. Sole say it muddies up the tone when you put 1K on a 6V6 of 6L6... but I challenge those people in a blind test.

Thanks for the insight on 470r versus 1k.

In example #2 in my photo, is it problematic to add the individual screen resistors while leaving the original wire that connected the screens both to r240 on one side of the 6aq5s and C216C on the other side of the 6aq5s in place? Does that create a loop that starts affecting resistances?

Maybe this is the simplest and most effective way to add screen resistance?

[Bieworm]

If you look at other 18W schematics , such as Ceriatone, then you'll notice they use a single 100R resistor from the filter cap to split into 2 wires to the screens after the resistor.
That's fine too I guess.
I put 1k on my 6L6 amps too, so I can put in EL34 when I feel like it. I don't think there's a lot of difference between 470R and 1k for the screens. Sole say it muddies up the tone when you put 1K on a 6V6 of 6L6... but I challenge those people in a blind test.

Thanks for the insight on 470r versus 1k.

In example #2 in my photo, is it problematic to add the individual screen resistors while leaving the original wire that connected the screens both to r240 on one side of the 6aq5s and C216C on the other side of the 6aq5s in place? Does that create a loop that starts affecting resistances?

Maybe this is the simplest and most effective way to add screen resistance?

You drew it wrong. R240 should be connected to C216 C , at the same connection the screen resistor is going to be added. Also, I would do the 100R instead of 1k when you add a single resistor to both screens, just like the Ceriatone 18W TMB.

[JMPGuitars]

I would do the 100R instead of 1k when you add a single resistor to both screens, just like the Ceriatone 18W TMB.

No. Do 1K. You can do 1K split by two 100 ohms to each tube if you want, but the simple version with 1K is fine.

[yello]

You drew it wrong. R240 should be connected to C216 C , at the same connection the screen resistor is going to be added. Also, I would do the 100R instead of 1k when you add a single resistor to both screens, just like the Ceriatone 18W TMB.

No. Do 1K. You can do 1K split by two 100 ohms to each tube if you want, but the simple version with 1K is fine.

So should it be like option number 1, in the middle of this picture? That way adding the 1k screen resistor is just a slight modification from the stock wiring which is pictured above it, and goes to the same point as R240. Or I could have R240 and the 1k both arrive at the right hand tube instead.

[Bieworm]

You drew it wrong. R240 should be connected to C216 C , at the same connection the screen resistor is going to be added. Also, I would do the 100R instead of 1k when you add a single resistor to both screens, just like the Ceriatone 18W TMB.

No. Do 1K. You can do 1K split by two 100 ohms to each tube if you want, but the simple version with 1K is fine.

So should it be like option number 1, in the middle of this picture? That way adding the 1k screen resistor is just a slight modification from the stock wiring which is pictured above it, and goes to the same point as R240. Or I could have R240 and the 1k both arrive at the right hand tube instead.

You still have R240 at the wrong connection. The screens should not be in series with that resistor.

Screenshot_20220423-222037_Chrome.jpg

[JMPGuitars]

screens.jpg

I dunno what the actual amp looks like now, but any one of those options are fine. The designations are based on the schematic, and what you wrote.

schem.jpg

[yello]

You still have R240 at the wrong connection. The screens should not be in series with that resistor. Screenshot_20220423-222037_Chrome.jpg

I dunno what the actual amp looks like now, but any one of those options are fine. The designations are based on the schematic, and what you wrote.

I'm happy to fix it according to your advice, I want this amp working properly.

The struggle I am having is how they were all wired from the factory (these are from various years of the mid 1960's). I have 4 of the amps currently, and all 4 of them have R240 going to the screen of V203. All 4 have a wire connecting v203 to v204. 3 of the 4 have C216C going to the screen of V203, same place where R240 is landing, while 1 of the 4 has C216C going to the screen of V204 instead.

Here is a diagram of what I see in all 4 amps, with R228 added so you can see how C216A and C216B are incorporated. The only variation is noted above, namely where C216C lands (noted in my diagram as either "X" or "Y".

Is this a wiring mistake the factory made over several years? Or did it serve some intentional purpose?

I've measured voltages in 2 of the 4 amps and the screen voltage in both was about 330 while plate voltage was about 320 - is this wiring mistake the culprit? If so, I assume it would be corrected by wiring per Josh's diagram, which is easy to implement.

This is how all 4 are wired from the factory:



And here are pics of three of them, its the orange wires that connect C216C, V203, V204, and R240:

[JMPGuitars]

R240 being soldered to V203 or C216 C is electronically the same in your diagram. The only question is what is the dashed Y line you drew?

Some amps are made with the screens slightly higher than the plates. This is not ideal for guitar amps.