Bright Bulb, Short, Flummoxed & Blue

[rp]

Intro: I've been retro-fitting all my homebrew amps with 120V/220V PTs and doing recapping / general maintenance. This 5E3 is my second ever build from many ages ago. It's been changed around many times, hence the mess. Started out all metal film and MIT caps in a dumb phase of my life. I added the Black Cats in atonement but they just made it a jumble of hell and I should have used small 150s and saved the cats for a proper 5E3 combo build w/ a board. The big Black Cats may still come out as well as the crappy blue bourns pots.

The problem: After adding new PT, electrolytics, and filament rewire (to change 120V light to 6V) I power up with my new bulb limiter, starts well enough - quick flash, then dims as it should as the e-caps charge up, rises to about 320V on the plates then immediately the bulb (40W) starts to brighten and voltage drops quickly away, about 15 seconds in all. (I guess that's the bulb magically dropping the voltage - cool, thing this.)

The bulb stays dim with just the rectifier and pre-amp tubes and the voltages are stable, but when the power tubes go in it brightens. Same if I try one power tube at a time. I tried different power tubes too, same thing. HT fuse doesn't blow but I'm not going to hazard plugging it in w/o the DBL.

Wiring looks ok, I traced it all twice, but I did mess around a lot but not around the preamp/PI. I checked pins of every tube and got Mega ohms to ground where i should. I had the amp lent out to a friend before I started all the new work. I don't remember if I checked it when I got it back. So maybe there was a previous underlying issue, I hope the OT is ok. I get 600 ohms across the primaries 300 ohm to the CT, meg ohms across the pri & sec, 1 ohm from sec to ground.

Maybe the new PT is wired wrong, the 120/220V is a series/parallel set up, but I followed the wiring directions from heyboer and I'm getting 6.3Vac on the filaments, 5Vac rec, 660Vac sec as spec'd. Must be OK?

I really don't want to shotgun this thing right away, I want to check out the Black Cats a bit more with the new PT and new more proper lower plate V. Maybe there's an easy answer and I'll learn some trouble shooting, or someone will spot the mistake.

http://picasaweb.google.com/ardpan/5E3C ... directlink

Sincerest apologies to anyone's eyes hurting looking through the clutter.

BIG LOVE for the bulb limiter - thanks Ampgaragers for pushing me to build one. My favorite kind of thing - simple, cheap and incredibly useful (keeps the smoke in) never goes obsolete (if incandescents don't) the one I built will last a lifetime. Now I want to make one with a 3-way bulb.

A free bulb limiter to anyone who spots the problem!

[labb]

Just a wild guess but it sounds like you have a problem in the cathode biasing section of the power tubes..All though I could not see a problem in the pictures. Also check for DC on the grids at pin 5 of the power tubes.

[rp]

No DC on control grids. Biasing looks correct to me too, I keep staring at it. How long can I leave this powered up with the lit DBL and the power tubes in? Few secs, few mins, as long as I want?

[Alexo]

Tried a different set of power tubes?

[rp]

yup, maybe I'll try a 3rd set.

Just did, no luck. I'm stuck. This is why I put everything in storage and stopped doing amps for 10 years. When I get really stuck like this I have no choice but to start ripping out and redoing whole stages or even the whole amp if I get really ticked off. I'll wait a few days and see where this post goes then I start ripping stuff out.

Have I definitely localized it to the output stage? The power supply & pre must be ok as it's stable w/o the power tubes, correct?

[Structo]

What do you have going on with the output jacks?
Is there a line level there or something?

You probably know this but any time you turn on a tube amp with power tubes installed, you need to have a speaker load connected.

Are you getting any sound during the brief moments you had it powered up?

It's hard to troubleshoot point to point pictures because of the 2D aspect and no depth.

From what you say, it appears to be the power section or the B+1 that is drawing too much current.

I didn't really follow when you said the bulb flashes then dims, then gets brighter again.
That shouldn't happen.
Once the caps are charged there is very little current draw.

The reason the bulb flashes is that when the switch is first thrown, the caps almost appear as a dead short for an instant before the charge builds in them.
But, the bulb should go dim and stay dim after they charge.

So check everything connected to the power tube sockets.
I'll bet it is something simple that you just aren't seeing and you will slap your forehead when you find it.

Best thing to do sometimes is walk away for a couple days.
Then come back fresh and you will see new things.

Good luck!

[Structo]

Looking at the pics some more.

Does the red wire from the cap can lug and 22K resistor go to V1 plate (100K) resistors?

Just making sure that the B+ are going where they should.

Does the OT center tap (red) go to the first node? (5K and 16uF)

[rp]

What do you have going on with the output jacks?
Is there a line level there or something?

Yes, would this interfere?

You probably know this but any time you turn on a tube amp with power tubes installed, you need to have a speaker load connected.

For the above test I had an 8ohm/50W R. Was too lazy to get a speaker. Bad idea I realized; it's good to hear things.

Are you getting any sound during the brief moments you had it powered up?

Just tried a speaker. Yes! I get hiss (the good kind) with pots all up. Amp settled to 160V plates with the DBL/40W bulb.

I didn't really follow when you said the bulb flashes then dims, then gets brighter again. That shouldn't happen. Once the caps are charged there is very little current draw. The reason the bulb flashes is that when the switch is first thrown, the caps almost appear as a dead short for an instant before the charge builds in them. But, the bulb should go dim and stay dim after they charge.

With the caps drained I power up and I get the brief flash of the bulb which i understand is from the uncharged caps, the bulb quickly dims and all looks good as the caps charge till I hit around 300V on the plates then B+ reverses and starts to drop quickly as the bulb starts to brighten.

Does the red wire from the cap can lug and 22K resistor go to V1 plate (100K) resistors?

Yup, see image of that node. Even my camera complained about going in there. Those Blacks Cats are going to go soon.

Does the OT center tap (red) go to the first node? (5K and 16uF)

Yes. Thanks for the time Structo. I think I'm on to something. BRB

[passfan]

Why is the first F/T filter cap after your standby switch not grounded, Am I missing this? I see positive to the terminal with the 5 watt cement wirewound but the terminal on the other end looks like it's going nowhere.

[rp]

Why is the first F/T filter cap after your standby switch not grounded, Am I missing this? I see positive to the terminal with the 5 watt cement wirewound but the terminal on the other end looks like it's going nowhere.

It's grounded along the bottom of the terminal strip to the bus bar, below the cap w/ a piece of solid core and Teflon tubing, like I did the heaters to the trany leads. I don't really like using the holes there, and i don't like hiding connections in case someone other than me ever works on it, but sometimes the neatnik in me takes over.

[rp]

I figured it out. I just don't know how to use a DBL. After Structo's post and I heard hiss I put it on my variac and went for it. It played and all voltages were normal. It was the 40W bulb messing me up. It glows a lot brighter than a 100W which is what I first used when I brought my Lightning up a few weeks ago. I didn't feel like taking a ceiling fixture apart for a 100W so I used a 40W I had at hand. Makes a big difference! The 40W looked darn bright to me, the 100W just had a dim filament glow. I figured it out when I tried the DBL/40W with all my amps and all were the same brightness. I knew the different wattages limited more/less, but I didn't know they glowed so differently. And the 40W really took my voltages down which threw me off.

Now what I don't understand is the progress the 40W made, Flashed on (got that, cold caps) then dimmed to just a glowing filament as the B+ ramped up to full, then started glowing much brighter as the voltage dropped. The 100W didn't do this, instead shot right up to 425V (seeing the uncharged caps) and then just progressed down to 250V with the bulb staying about equally dim the whole time.

Guess with this old technology you just gotta develop a feel for it. Thanks all, sorry for the false alarm.

[RevD]

I might be wrong, but I think the lamp limiter dealeo is more for just determining if your having a problem drawing to much current on a new build. After you determined it doesn't have a dead short and nothing blows, it was my recollection that you then just plug it in direct and tweak from that point?

Regards,

Rev. D.

[Structo]

Yeah, I don't get that bulb brightening back up either.

On a brand new build or recap I use a 25w bulb in my limiter.
First with no tubes.

This limits the current to less than 250ma.

E/R=I

It also forms the caps more slowly than just nailing them with the full 120vac/15A supply.

If the amp does OK on that test.
Then I install the tubes, put a 60w light bulb in and repeat the test.
If there is no short or high current fault I leave that on for a while.

Turn it off, wait a while to let it cool then power it up straight from the wall.
IF the amp is fixed bias I would have already turned the bias pot for the lowest negative voltage before turning it on.
Then I bias the amp.
On a cathode bias I check what the power tubes are drawing and adjust that if need be.

The light bulb limiter is a great tool and can save the expensive stuff and keep the smoke where it belongs.