"Fiddy" 50-watt build log

[xtian]

Here are a few photos. Note the blinkM <http://blinkm.thingm.com> tri-color, programmable LED used as the power indicator. You program this little smart light (I used an Arduino) then it remembers its program even when unplugged. You just need to supply 5vdc and it will run its program, which I supplied with the unused 5vac rectifier supply, a diode, resistor, and capacitor to smooth the dc.

It's impossible to capture the colorful nature of the blinkM with a consumer camera, so I made a stop motion movie with my Canon 30D, with fixed focus and exposure, then assembled in QuickTime:

http://youtu.be/gg72VGxTO-0

[xtian]

More photos.

[Reeltarded]

You kids and your pesky LEDs.

nice amp dooder!

[xtian]

Stress test = blow out! Whoo! Drove my amp up the hill to play loud with my buddy. Started to dime this beast in cascade mode, and, FLASH! Got a nice blue arc light, lasted almost a full second!

Turns out to be both of the 100ohm resistors acting as heater center tap. See photo. Ohms Law says 6.3v / 100ohms = 0.63a * 6.3v = 0.4 watts. And I'm using half-watt resistors, which appears to be too close for comfort.

None of my close-quarters wiring arced, so I don't think that's an issue. Anything else I should check before I upgrade to 1-watt resistors and try again?

Also, if hum is not an issue, can I omit the artificial center tap altogether?

[ToneMerc]

Turns out to be both of the 100ohm resistors acting as heater center tap. See photo. Ohms Law says 6.3v / 100ohms = 0.63a * 6.3v = 0.4 watts. And I'm using half-watt resistors, which appears to be too close for comfort.

6.3/100 = .063

Unless I'm wrong, because of the artificial CT it's 3.15V/100 ohms = 32mA for each resistor.

I would be more concerned about the integrity of that PT.

TM

[ChrisM]

Turns out to be both of the 100ohm resistors acting as heater center tap. See photo. Ohms Law says 6.3v / 100ohms = 0.63a * 6.3v = 0.4 watts. And I'm using half-watt resistors, which appears to be too close for comfort.

6.3/100 = .063

Unless I'm wrong, because of the artificial CT it's 3.15V/100 ohms = 32mA for each resistor.

I would be more concerned about the integrity of that PT.

TM

+1
Those resistor should be seeing very little current go through them.

Check for shorts.

[Phil_S]

Someone tell me if I'm not right about this. Heater current should be constant. It doesn't change when you dime the amp. This, along with the estimate of 32mA per resistor suggests there is some other problem here.

When you dime the amp and run it hard, it can stress the PT. IMO, you've got to look elsewhere for the real problem.

OTOH, is there anything close by that may have arced? Are there any unintended grounds to the chassis near the filament artificial CT? Did you make the connection correctly?

I can't see how you actually made the artificial CT. This needs to be two 100R made into a "V" shaped joint (or the circuit equivalent) with the point of the V grounded to the chassis. This should NOT be part of the circuit ground. It should be separate. You can put a ring lug under the nut near pin #9 and terminate the artificial CT there.

[xtian]

Thanks, Phil, heater CT is correct--one 100R from each leg of the heater to ground (they are grounded on the center lugs of those tag strips, which are firmly bolted to the chassis, and nothing else is grounded there.

When I repair this, I'll disconnect the heater wires at the PT, and check the heater runs for shorts to ground.

What else should I check? The amp was running perfectly for a few practice sessions at home.

[xtian]

Just heard from my buddy who's cab we we're using…it's a 16 ohm cab. I have a 2-ohm output. Duh!

Can you imagine how such a mismatch might stress the PT and fry the heater center tap?

[ToneMerc]

Just heard from my buddy who's cab we we're using…it's a 16 ohm cab. I have a 2-ohm output. Duh!

Can you imagine how such a mismatch might stress the PT and fry the heater center tap?

Some degree of an intermttent(maybe) short facing the secondary windings of that PT is my guess. It could be anything from PT windings to a bad power tube. The mains fuse didn't open right?

Divide and conquer

TM

[martin manning]

An 8x mismatch on the load resistance would generate some very high voltages on the OT primary. Maybe the arcing was from the plate to the filament circuit somewhere, with the path to ground being the artificial CT resistors? Replace those and hope that the OT has not been damaged.

[Structo]

going up two steps in speaker impedance can cause flyback voltage, correct?
Hopefully you didn't cook the OT.

Just to be clear, you didn't have the PT heater center tap and the 100R resistors connected, right?

One or the other is correct.

About a year ago I was check voltages in my 100w amp and was measuring the 6L6GC plate voltage.
Well my hand was shaky and I slipped off pin 3 and shorted it to pin 2 (heater).

Sparks flew and there was a puff of smoke.

Near as I can tell, I just fried the two 100R center tap resistors.
They measured open.

Replaced those and it was fine.

BUT, I learned a valuable lesson.
If your hands are shaking use clips on the test leads and connect them before you energize the amp.

[xtian]

Thanks, Martin and Tom. I discovered that the HT fuse (500ma) was burnt, so that failsafe operated properly. No other damage, other than the 100R CT resistors. I have installed new ones (150R for slightly lower wattage requirement) and powered up without tubes to check the PT, which seems fine. I don't have any 500ma fuses, so there will be a break in the action until I get new ones and am able to check the OT.

Tom, there is no native CT for the heater supply.

[Cameron]

Someone tell me if I'm not right about this. Heater current should be constant. It doesn't change when you dime the amp. This, along with the estimate of 32mA per resistor suggests there is some other problem here.

When you dime the amp and run it hard, it can stress the PT. IMO, you've got to look elsewhere for the real problem.

OTOH, is there anything close by that may have arced? Are there any unintended grounds to the chassis near the filament artificial CT? Did you make the connection correctly?

I can't see how you actually made the artificial CT. This needs to be two 100R made into a "V" shaped joint (or the circuit equivalent) with the point of the V grounded to the chassis. This should NOT be part of the circuit ground. It should be separate. You can put a ring lug under the nut near pin #9 and terminate the artificial CT there.

Don't put filament supply parts near hi voltage...if something goes wrong it WILL short to the filament supply. The problem is just a bad tube....IMO.

[amplifiednation]

Nice looking build. Its a sweet thing when the house is empty isn't it? Its even better when the house is empty, the neighbors houses are empty and its raining. THEN is time to rock! I'm curious if you keep the overwhelming gain? Did you use a cathode follower to drive the tone circuit? Sorry, Im gonna need to look at the layout again