Supro S6606 build

[AnalogDok]

I'm at it again. I decided to do the Supro S6606 because it's different than all the Tweed Princetons out there. Similar but different. I have a nice Weber 8" speaker laying around so I decided to build a combo amp.
I'm running into a problem creating the turret/eyelet board layout. I get going on the drawing and then it quickly gets messy and confusing. It's not the same as the 5F2A but it's close in some areas. I've been looking for another layout that matches the schematic but failed so far.
I haven't found any software to do it as they are all focused on digital projects and PCB production.

Does anyone have a layout or can help me out?

Thanks

[sluckey]

Try this...

     http://sluckeyamps.com/misc/Supro_s6606.pdf

[AnalogDok]

Holy Heck....Wow...

I'm looking over the layout vs the schematic and I'm learning. I'm starting to understand the way to do this.
I spent 4 hours of crumpling paper, knowing I was not right. It's a learning curve and I sure appreciate the lesson here.
I see where you changed the position of the .01 cap and Tone pot to the other side. Smart.

I have a spare Valve Jr OT and a Weber Champ PT to do this build. Components are here too. I'm waiting on the blackened tuna tolex and grill cloth and the pine box to build the cab.

Thank You

[AnalogDok]

I built a finger jointed pine box and covered it. Took a blank 5E3 chassis and chopped the height down. Drilled it and mounted all the hardware to check the fit. I made the baffle out of 1/4" Baltic Birch plywood and glued on the grill cloth so I didn't destroy the baffle with staples.. cut some vinyl and fitted some bakelite witch hats. All the parts are soldered onto the board.
I can't thank Steve Lucky enough. He has been my inspiration on the last 3 builds.

[sluckey]

Looks like a Supro to me. Is it finished?

[AnalogDok]

Not yet, I'm doing the S6606 and the 5E3 at the same time. Hopefully in the next few days if the "Honey Do's" aren't too numerous.
I fitted the faceplate to the 5E3 today and plan on installing the S6606 main board tomorrow.

[AnalogDok]

I fired her up tonight. She behaves perfectly from 1 to 10 on the volume. Exactly as I expected.
But.....
If I turn up any tone at all she starts squeeling and then begins to motorboat at about 8 or so.
I checked the wiring and verified it all is correct and soldered properly.
I moved wire around with chopsticks but no change. Something isn't right...

I decided to take a break and look again tomorrow.
Great sounding amp tho..No hiss, hum or noise at all at any volume.

[PRR]

> she starts squeeling

You have the very small input signal wires real near your large power plate wires. It begs for sneak-back and oscillation.

If you don't have even an approximate Supro layout, plagiarize a Fender. This probably means putting the 12AX7 parts over near the 12AX7, far away from the 6L6/6V6 parts. (Yes, about where that extra 1/4 jack is.

[AnalogDok]

Thank You PRR,

Sometimes it takes someone else to say what you know already.
I did my usual fix. It worked fantastic on a JTM50 Black Flag where I moved some stuff around.

I just did a test and quiet as a mouse.

Thanks.

[sluckey]

I see you even "shielded" the ground wire to the tube. But shielding is useless if you don't connect the shield braid to anything. Your fix is simply due to rerouting the wires.

[AnalogDok]

You can't see the ground wire from this angle. I wrapped all the shield cable grounds together and soldered them to a ground lug bolted through the chassis near the 12AX7. The lug is right underneath the plastic tie wrap. The tone and volume grounds are connected to the main ground run across the pots. I't a shield ground so there should be no ground loop issues. All my other grounds are in one star ground spot near the PWR Tranny.

I ran the amp this morning and at low volume, it's fine but the squeel starts up when you go past 3. It also has a farty crackle I didn't hear before, It's very pronounced when the volume is past 3.

I'll do some voltage measurements and try to trace it with an oscilloscope later today.

[AnalogDok]

Here are my measurements. Fluke 77

Line in 124.4 VAC Transformer set to 125 VAC
Heater is at 6.58 VAC under load

V1 (12AX7)
Pin1  134.1 VDC
PIN2  .587 VDC
PIN6  160 VDC
PIN8  1.56 VDC

V2 (6V6)
PIN3  364 VDC
PIN4  345 VDC
PIN5  .015 VDC
PIN8  18.9 VDC

Points as marked on the layout
A = 379 VDC
B = 347 VDC
C = 267 VDC

[sluckey]

Your layout is the problem. I don't think you can stop the squeal with that layout.

[PRR]

Not just the wires. The caps and resistors on the board are also "antennas".

Fender would put most of that board over near the 12AX7; a longer board so the 6V6 stuff can go on the other end away from the little stuff.

[AnalogDok]

PRR and slucky,
I was thinking of routing the 6V6 wires under the board to separate the wires but a longer board and moving the 12AX7 parts away from the filters and 6V6 sounds reasonable. If at first you don't succeed, try, try again. I have some more board and turrets.

I have a Tweed Princeton layout in front of me and that's how Fender did it.

[sluckey]

My board layout was designed to be used in a chassis that was laid out like my layout drawing. The board must work together with all the other components. Tubes should have been laid out as in my drawing. Same with front controls and inputs. But as logical as my board layout is, it just won't work well with the way you laid out your chassis. I think you will be time ahead to just ditch my board and start over, maybe even just build true PTP with terminal strips. But in any case... MOVE THAT SPEAKER JACK FAR AWAY FROM THAT PREAMP TUBE. AND SHORTEN THOSE INPUT WIRES.

[PRR]

+1 on speaker jack; but 6V6 Plate wires are 30 times more radiation and sneak-back to low level stages.

[AnalogDok]

I laid out the parts a bit different.
I think this will work much better. I ran some beryllium copper shielding around the chassis to see where the radiation was coming from and the speaker jack wasn't the issue. Since I moved all the preamp tube signal wires and shielded them it had no effect there either. Laying the parts out with separation both for distance and function should resolve the radiation issue on the board.

[PRR]

Not the same amp, but a comparable side by side: look OK. Worked for Leo.

[shooter]

beryllium

as the new guy, the old timers showed me how to dawn a level III bio-suit so I could remove beryllium oxide filters used to shape Xray beams.  the FDA figured out the "dust" was really bad for human consumption.  the overtime money was worth the sweatsuit   

[AnalogDok]

They make screen rooms out of beryllium copper screen. It blocks all RFI radiation so we can test electronic devices for FCC certification.
It allows an RF Engineer to isolate the device from any outside radiation/RF Interference.

[AnalogDok]

PRR,

That 5f2-A was my goto example to resolve the crossed signals in the last board. Like I said I did some shielding tests and isolated the board as the main culprit. My layout was the same as Steve's but had the board shifted away from V1. That was a bad placement but not the whole problem. Having the preamp section wedged in between the 10uF cap and the V2 circuit needed to be addressed. I thought about it but just built it that way and it didn't work out. My bad...this is an exercise in learning and Steve's layout taught me a lot how to convert a schematic to a layout.
I completely missed the resistor on PIN 8 V1 BTW...I made a ton of mistakes to get it this far but I'm learning and that is the whole reason for this build. I've gained a ton of new skill doing this. Very happy...

[shooter]

one of the prettiest things i ever seen was the morning I walked into a new MRI room after the RF guys put up the copper walls.  our "cages" needed to be like 120db down from the fundamental frequency out 250khz either side.

[AnalogDok]

Board Ver 2...

I verified all the parts were within spec and used a heat sink when soldering.
Now to install and test.

[AnalogDok]

I'm shocked...it worked...zero squeal no matter where I set the volume and tone.
It does squeal like a banshee if no guitar is plugged in tho and there is a slight buzz.
Next is isolating the the reason for the squeal with empty input jacks and then the very slight buzz.

It's bright, doesn't sound like a Tweed Princeton and the distortion starts after 7 on the volume knob.
It doesn't sound remotely like a Fender in distortion. It sounds really good on 10...

[sluckey]

Next is isolating the the reason for the squeal with empty input jacks and then the very slight buzz.

That's easy. Use Switchcraft 12A switching jacks and wire them IAW the schematic and the layout I provided.

[AnalogDok]

So when nothing is plugged into the input, the jacks are automatically grounded. Yeah, simple solution.
The amp sounds the same as before, just no squeal now.

I'm gonna stuff a 6n2P tube in it with an adapter next.

[AnalogDok]

The jacks worked great and the 6n2P has a little more headroom, compression and smoother distortion.
I do hear some artifacts when it is cranked so I'll try some other 6V6 tubes. I'm running a Motorola from the 1960's at present.
I went in with a chopstick and touched everything with the guitar hooked up and everything is solid and moving wires around does absolutely nothing.

[AnalogDok]

Special thanks to Slucky.

You are the inspiration and the only reason this amp is alive.
I replaced the ancient Motorola 6V6GT with a brand new JJ 6V6S I had laying around and there is no after effect or artifacts at all. It's just perfect now. New jacks installed completely eliminated that squeal when nothing was plugged into the input jack.
I moved the output jack to the only available area which was in between the transformers. The chassis is a 5E3 so it's 14.5" long X 3.5" wide and the 3.12" board barely fits and I only have 3/4" standoffs so there wasn't much choice except to move towards the transformers opposite the preamp tube.

It all works exactly as it should. The Weber 8" speaker is a bit bright but then 8" speakers are limited anyway.
The OD doesn't sound like Fender at all. I like it and it doesn't get mushy. It sounds pretty much like a S1606 like on the Youtube videos.

The only thing I'd like to have is a HI and LO input instead of the same dual input....maybe...

[sluckey]

The only thing I'd like to have is a HI and LO input instead of the same dual input....maybe...

That's easy too!

[AnalogDok]

So both my inputs stock are Hi.

I was trying to figure out how a 1M resistor was going to affect my sound...So simple. I was overthinking again as usual.

All I had was 2 watt resistors. Is that going to cause any problems and should I change them to 3 watt?

[sluckey]

You don't need any extra resistors. All you have to do is remove the jumper on the switch lug of your LO jack. Now connect a wire from the LO switch lug to the HI tip lug. That's all.

Look closely at the new layout I posted.

[AnalogDok]

I meant the 2 watt resistors were used on the filter caps instead of the 3 watt on your layout.

[sluckey]

I meant the 2 watt resistors were used on the filter caps instead of the 3 watt on your layout.

ARRGH! RED HERRING! WE WERE DISCUSSING INPUT JACKS!   

[AnalogDok]

Sorry about that Sluckey,

So the question is did I under build the filter section by using 2 watt resistors?
Should I change them out to 3 watt?

[sluckey]

Sorry about that Sluckey,

So the question is did I under build the filter section by using 2 watt resistors?
Should I change them out to 3 watt?

Your resistors are fine. Supro used 1/2 watt for those. I use 3 watt metal oxide for all resistors on the B+ rail because they are flameproof and Hoffman sells them. Overkill in most cases but I like it. When that original schematic was drawn there were no 3 watt metal oxide resistors.

Do your HI/LO jacks work?

[AnalogDok]

I put the mods and final inspection of the S6606 on hold while I finish my 5E3 project.

[AnalogDok]

Here is a shot at the amp's interior.
I like to use 10 Watt Dale wirewound resistors in place of the 5 watt and 3 watt in place of the 1 or 2 watt.

It just adds safety. None of my amps have Chinese made Xircon.

[AnalogDok]

I did the HI_LO input mod tonight. Worked like a charm Steve.

Thank You.

[K Teacher]

Hi AnalogDok,

Excellent post… very detailed with good pics and voltage values.
It is very instructive for anyone that reads it… thanks.

You may or may not know this, but based on the voltages you posted, the 6V6 is running at approx. 18W of plate dissipation and this will shorten its useful life.

As a suggestion to remedy this, you might wanna change the PT for one like the Hammond 290HAX or replace the 6V6 330-ohm cathode bias resistor for 560-ohm / 5W.

Just my 2 cents…

[PRR]

> based on the voltages you posted, the 6V6 is running at approx. 18W of plate dissipation

Which numbers are you looking at?

Reply #11 seems to say 12.05 Watts in plate.

[sluckey]

Ignoring screen current I calculate 19.8 watts.

[K Teacher]

Using the numbers from reply # 11:

Cathode_current (Ik) = Cathode_voltage (Ek) / Rk
Ik = 1,000 x 18.9 V (6V6-pin8) / 330 ohm
Ik = 53.3 mA

Plate_current (Ib) = Cathode_current (Ik) - Screen_current (Is)
Using 6V6 tube manual and calculated Ik, we get:
Ib = 52.1 mA

Plate_dissipation (Wb) = [Plate_voltage (Eb = 6V6-pin3) - Cathode_voltage (Ek)] x Plate_current (Ib)
Wb = (364 - 18.9) x 52.1 / 1,000
Wb = 345.1 x 0.0521
Wb = 18.0 W

[PRR]

Ah, are we then using 330Ω for Rk? Then yes, hot.

[66Strat]

Using the numbers from reply # 11:

Cathode_current (Ik) = Cathode_voltage (Ek) / Rk
Ik = 1,000 x 18.9 V (6V6-pin8) / 330 ohm
Ik = 53.3 mA

Plate_current (Ib) = Cathode_current (Ik) - Screen_current (Is)
Using 6V6 tube manual and calculated Ik, we get:
Ib = 52.1 mA

Plate_dissipation (Wb) = [Plate_voltage (Eb = 6V6-pin3) - Cathode_voltage (Ek)] x Plate_current (Ib)
Wb = (364 - 18.9) x 52.1 / 1,000
Wb = 345.1 x 0.0521
Wb = 18.0 W

I get the same number as sluckey, 19.8 watts.  I think you may have erred in your cathode current calculation. I get 57.27 milliamps, (1000 x 18.9 / 330 = 57.27) when multiplied by plate to cathode voltage of 345.1 (364 - 18.9) correlates to 19.8 watts. As Steve indicated, this ignores screen current. Allowing for screen current of ~2 milliamps, plate dissipation would be ~19.1 watts (345.1 x .05527).

Regardless of whether it's 18 or 19.8, it's still way too hot for a 6V6, and your recommendation to increase the cathode resistor is spot on.

[K Teacher]

@66Strat,

Thanks for pointing out this typo...
I mistakenly typed 53.3 mA instead of 57.3 mA (57.27 mA IS 57.3 rounded).

However, for the values indicated in this thread, the 6V6 plate current is still 52.1 mA and plate dissipation is still 18.0 W.

You and Mr. sluckey got 19-point-some-watts of plate dissipation, because you did NOT subtract screen current (Is) from cathode current on your calculations…
OR you estimate too low screen current... check values in the 6V6 datasheet...

[AnalogDok]

Thank you K-Teacher for pointing this issue out.

Other than the initial test and a couple of short sessions, I haven't used the amp much. It is louder than it should be and with the little Weber Signature 8" speaker, it overpowers it very quickly. It sounded great.

The Bias was at 54mA

I pulled it out to test it tonight after reading the posts and learned 2 lessons tonight.
#1...If the alignment lug on a rectifier tube is broken off, toss it out.
#2...Always verify the tube bias on all cathode biased amps. Do not just assume it is correct because you followed the schematic.

First off, the schematic calls for a 330Ω resistor to cathode bias the amp. I have a 5E3 that I used Hammond A029 trannys on and I had to change the cathode resistor to bring it within spec due to the higher voltages. I should have checked the S6606 too, especially when it was so loud.

I pulled out this old EH 5Y3 that I used in the build and the 6V6 to flip the amp over to remove the 330Ω resistor and when I reinstalled the tubes, I assumed the keyway was on the tubes and inserted them without looking. I was going to test with several different values of 5 and 10 watt resistors I have in my parts box and was making sure the aligator clips were connected properly and wasn't paying enough attention to the 5Y3 as I totally forgot it had a broken keyway. I fried the power transformer. The 2A fuse blew way too late.

So while it may seem OK to visually align a keyless rectifier tube while testing an amp, it ended up costing me more than double what it was worth by leaving it in a finished build and forgetting about it's danger. It was the only 5Y3 I had left.

I ordered a new power transformer tonight, 2 new 5Y3 rectifier tubes so I have a spare and some Ohmite wirewound resistors. I ordered 10 watt resistors in 560Ω, 562Ω and 600Ω for options. About $125.00 total.

While the new Chihuahua puppy was chewing my fingers, mouse and keyboard, I updated the DIYLC layout with the current numbers when it had the 330Ω resistor in place. Once I install the new transformer, I'll update the DIYLC layout and verify the Bias.

[sluckey]

You may be interested in these...

     https://www.tubesandmore.com/products/keyway-replacement-piece-octal-tubes

[Willabe]

Modern made 5Y3 rectifier tubes, and your EH is modern made, will give higher dcv than NOS 5Y3's.

[AnalogDok]

Thanks guys.
Sluckey, Those Keyway replacements look cool but the plate looks to be about 1/8" thick and that loss of insertion into the socket may have a negative effect. I think the actual pin insertion is maybe 1/4" so the pins will be barely connected at 1/8". I'd be more inclined if the top plate was thinner.
$18.00 to replace the tube isn't much.

Willabe,
Spending $80.00 on a NOS rectifier is not going to happen. I have a Variac for that.
I don't think I have any voltage issues with a standard Champ PS.