New Amp: Purple Flashes, No Sound

[The New Steve H]

Absolutely not. I used a crummy light I had lying around, and I didn't pay much attention to where I was sticking it. No point in lying about it.

It's a 125V light. Let me see if I can find a link to it.

http://www.mouser.com/ProductDetail/Ch ... UsCw%3D%3D

I think I know how to fix it, but I don't want to post my idea until after the coffee takes effect, because I know what usually happens when I do that.

Okay, I'll blab it. I think I should put it after the main switch, in parallel with the whole rest of the amp. I wonder if that will still make sense half an hour from now.

It has become obvious that I know virtually nothing about power supplies, and I have also learned that real schematics (unlike teaching materials) leave out details because they assume you're not a complete doofus. I broke down and ordered Merlin Blencowe's OTHER book.

[The New Steve H]

I stuck a .47 uF capacitor between ground and my standby switch, because I could not find my .047 uF capacitor. I am wondering: will the larger capacitor pass AC, causing the amp to draw too much current? I never thought about this when I was considering the problem. The capacitor goes straight to ground, so if current can get through it, there won't be a whole lot of impedance.

I know the formulas for high-pass filters and low-pass filters, but I have no idea how to apply them in this case. Do I just think of the amp as a resistor connected to the capacitor?

[Structo]

It will bleed any ripple to ground.

[The New Steve H]

I'm pretty sure I have all the wiring right, and I even stuck the correct capacitor in there and rewired the standby light. Now the amp doesn't blow the fuse, and it makes a really faint humming sound if you stick your head up to it. No output, however, and the rectifier still gives those flashes when I install it.

I turned it on with the guts visible, and I didn't see anything out of the ordinary. The flames were only shooting out about three or four feet.

I kid.

I am no rectifier genius, but it seems to me that if the rectifier is doing that, something must be generating sporadic current sucks on it, and that sounds like a short. I assume it's not normal for a rectifier to flash. And it only does it after it's been in there for a few minutes.

I will go back to checking voltages as soon as I get a new battery for my Fluke meter.

[Gibsonman63]

I would seriously recommend building a light bulb current limiter and using it until you get things sorted out. Mine paid for itself just in the cost of fuses.

I was reading through the tread, but I can't tell if you checked it with all of the tubes removed or just the rectifier. If not pull all of the tubes, put it on the current limiter and start from there.

Here's some ideas.

Check to see how much AC ripple you have on your B+
Double check the polarity of all the electrolytics
Check the plate voltage on all of your tube sockets.

Once you get B+ sorted out, check your power tube bias supply voltage. It's a half-wave rectifyer producing a negative bias voltage. Double-check your 1ohm resistors to ground on the power tubes. If you leave that one off, the power tubes won't turn on. (learned it the hard way clipping caps out of a silverfaced Fender and forgetting to put in the ground)

Pics would help. We have some eagle-eyed guys here that are pretty good at spotting things. Also, I have noticed on my camera when I set it for close-up photos, I can see things on the computer screen that I can't see on the amp even with a magnifyer.

1+ on Merlin's book. He gets over my head pretty quick, but he has some really good information on rectifiers. I'm about due another read myself to see if I can make a little more of it stick.

[The New Steve H]

I am just starting to realize how poorly I understand the power supply. I read up on triodes and tetrodes and pentodes, but it wasn't until about an hour ago that I really looked at rectifiers. I just wired this one up on faith. I didn't realize where the B+ came from, or that the diodes won't back up the rectifier unless you leave it in the socket.

I know the juice is getting to the rectifier, and everything up to that point is wired right, but I don't want to put the rectifier in until I get the shorting fixed. I am hoping the rectifier is okay, and I don't want to blow it up.

I guess I could put some temporary jumpers in the socket so I can run the amp without the tube.

I should have built that stupid current limiter already. I guess it's time for a ride to Home Depot.

Thanks for the help. I'll be printing out people's suggestions and taking them with me to the amp.

[The New Steve H]

The current limiter is built. It glows dimly until you flip the standby switch, and then it gradually gets brighter, and then it dims and flashes. The fuse doesn't blow, and the rectifier isn't going nuts any more.

Tomorrow I'll start checking voltages.

[The New Steve H]

Here is what I have so far.

I plugged the amp into the limiter. When I turned on the standby switch, I got a small brightening. I turned the main switch on, and the bulb slowly brightened and then began brightening and dimming.

I removed the tubes one at a time, starting with the greatest distance from the power transformer: preamp, phase inverter, output, rectifier. Nothing happened until I pulled the rectifier tube; the flashing (bulb, not tube) continued. When the rectifier tube came out, the bulb went dark.

I am guessing this means I have to look at what happens right after the rectifier. Maybe one of the caps is shorting?

[Gibsonman63]

Actually, it is just the opposite. The light bulb limiter puts the amp in series with the light bulb. If you have a direct short, the light bulb will shine at full intensity, the same as it would with a jumper wire in the circuit.

Since the bulb is completely dark, you have an open circuit... or the light bulb limiter isn't wired correctly.

[The New Steve H]

I am not totally sure if I was clear. The bulb is bright when the rectifier is working and dim when I remove the rectifier.

I found out what the biggest problem is. Elves. They broke in during the night and wired my power supply caps so the leads were too close to the cases. That explains where all my juice was going. Usually I have the yard sprayed for elves about once a year, but this damn recession...

Now the amp is running, but it's the world's biggest bedroom amp. The sound is very quiet, and it's a little distorted. Like listening to Otis Rush through a brick wall.

Incidentally, I DID remember to take it off the current limiter in order to check the volume issue. It got somewhat louder, but still about like a 1/20 watt amp.

Fun discovery: the volume knobs work backward. I assume the other pots are the same way. I had the layout diagram upside down when I wired everything, and I guess I followed it a little too closely.

Voltage at two main power caps: 367
Next cap: 367
Next one: 325

When I shut down the amp, the voltage in the power caps drops very quickly, so I think there is still a drain somewhere, unless this is normal.

[The New Steve H]

I fiddled around and re-checked the voltages on the power supply caps. Now it's 471 on all of them.

I'm getting more distortion than I thought. I assume this means the preamp is probably okay and the output section is the problem. Reasonable guess?

[Firestorm]

If I understand the way you have your SS diodes wired (1N4077s between the HT AC leads from the PT -- red wires -- and the plates of the rectifier tube), that is just a way of putting only positive voltage on the tube plates and preventing the tube from seeing any inverse voltage. It's a little nicer to the tube. But since the B+ still comes off Pin 8 of the tube (your yellow 5V wires should be on Pins 8 and 2, right?), so pulling the tube is kind of like opening the standby switch: it turns off the B+ connection to the rest of the amp.

Regardless, your heaters should still be powered, so the bulb should not be totally dark as long as the power switch is on. What wattage bulb are you using? 100W is typical.

What bothers me is the bulb "flashing" especially since you were getting purple fireworks from the rectifier tube before. Make very sure you've got the rectifier socket wired correctly. If all is good, I have to think the rectifier may be shot. Maybe you have to fess up and tell us how you had it miswired before so we can guess what bad thing might have happened.

EDIT: ahh, I see more more news came while I was typing. Picture time, I think.

[The New Steve H]

I can't say the bulb was totally dark. I know the heaters work, because I can see them glowing.

I think the rectifier is okay, and that the caps were shorting. If I understand the diode deal correctly, the amp should get DC now even if the rectifier tube is dead.

It turns out I didn't have the rectifier miswired, crazy as it sounds. I "fixed" it, then realized it was right the first time around.

[Firestorm]

When I shut down the amp, the voltage in the power caps drops very quickly, so I think there is still a drain somewhere, unless this is normal.

Is your main filter series caps with bleeder resistors?

Do you get 471VDC at the caps with tubes in or out?

[Firestorm]

I think the rectifier is okay, and that the caps were shorting. If I understand the diode deal correctly, the amp should get DC now even if the rectifier tube is dead.

With SS diodes "in front of" the tube, the DC stays up if the rectifier shorts heater to cathode. But not if the tube is pulled.