Homemade Tube amp blowing fuses

[sa230e]

Hey everyone,

So back in September I completed my first tube amp and I've been in love with it ever since I put about a year and a half of work into designing and building it and to my surprise and delight I got sound out of it the first time I plugged a guitar into it!  The tone is pretty decent for a first try and I've been playing it and tweaking it ever since. Once I got I finished tweaking I planned to make a youtube video to show it off but instead I started having problems with it in December. 

After a longer session I noticed the sound faded right out once or twice. I thought it might just be blocking distortion (although it didn't sound farty) but as I couldn't find a problem with it, I continued playing it. Then one day I switched it on and the mains fuse immediately blew.

So I pulled the power tubes, replaced the fuse and tried it again. Again the fuse blew.
So I pulled all the tubes, blew another fuse.

After hours of probing with a multimeter and not finding any shorts I started to worry about my power transformer. So I desoldered all the secondaries and terminated them (except for the HT which I can disconnect with standby switch) and tried it. The fuse survived.

So I resoldered everything, double checking for shorts and this time ran the amp through a dim bulb current limiter. It powered up. I plugged a guitar in and it sounded just like it used to.

Then I tried it straight into the mains without the dim bulb and immediately blew the fuse.

I don't know what's going on. I've gone through half a dozen fuses now trying to test with only one winding. It seems like it blows it with any one winding connected. Is my power transformer screwed? I have a slight temptation to pop in a larger fuse and see if anything gets crispy but I can't afford to fry that transformer as there's no way I'll get another one like that.

Here is a link to the schematics: https://dl.dropboxusercontent.com/u/3421497/miniplex.pdf

Sorry but they're hand drawn. I haven't found a program I like enough to make a more formal schematic. It's F1 that keeps blowing btw. It's worth noting that I never blew any fuses on the secondaries, always the primary fuse.

[Enzo]

On the bulb, with the transformer secondaries all disconnected, what does the bulb do?

[sa230e]

Nothing. It doesn't light up.

[g1]

Am I correct that it only blows the fuse when the standby switch is closed with the HT windings connected (even without tubes)?
If so, and it works with bulb limiter but not full line voltage, then the main filter cap may be breaking down at full voltage.

[sa230e]

Am I correct that it only blows the fuse when the standby switch is closed with the HT windings connected (even without tubes)?
If so, and it works with bulb limiter but not full line voltage, then the main filter cap may be breaking down at full voltage.

No, the fuse blows when the standby switch is open or closed. The HT fuse (F2) has also not blown at any time.

[Enzo]

First some clarity.  You CANNOT disconnect the HV wires with the standby switch.  Disconnect the wires and see if it matters.  I say that because the wires themselves could have a dot of solder shorting one. The switch opens the circuit to further in, but the switch could be faulty or the leads to it grounded somehow.  The HV fuse never blew, but that doesn't mean the holder has no shorts to frame.  The HV winding may not be involved, but it could be, and in ways that the standby switch won't affect.

You say it blows with one winding on, but can I assume that one winding also includes the switched off HV?  If so, see above.  So disconnect the HV wires.  Does it blow fuses with NONE of the secondary wires connected to ANYTHING?

I have a hard time wrapping my head around the fuses blowing, but the bulb not lighting.  What wattage bulb are you using?  I'd start with a 60 watt, after all, we are only powering the transformer at this point.

And with ALL the transformer wires disconected from everything, go to RG Keens geofex.com web page and find the transformer tester.  A very simple tool you can make with a neon bulb and a battery.

[sa230e]

First some clarity.  You CANNOT disconnect the HV wires with the standby switch.  Disconnect the wires and see if it matters.  I say that because the wires themselves could have a dot of solder shorting one. The switch opens the circuit to further in, but the switch could be faulty or the leads to it grounded somehow.  The HV fuse never blew, but that doesn't mean the holder has no shorts to frame.  The HV winding may not be involved, but it could be, and in ways that the standby switch won't affect.

You say it blows with one winding on, but can I assume that one winding also includes the switched off HV?  If so, see above.  So disconnect the HV wires.  Does it blow fuses with NONE of the secondary wires connected to ANYTHING?

I have a hard time wrapping my head around the fuses blowing, but the bulb not lighting.  What wattage bulb are you using?  I'd start with a 60 watt, after all, we are only powering the transformer at this point.

And with ALL the transformer wires disconected from everything, go to RG Keens geofex.com web page and find the transformer tester.  A very simple tool you can make with a neon bulb and a battery.

Fair enough, I'll desolder the HT tomorrow and let you know how it goes. It's 1:30 in the morning here.

Just to clarify a bit, the mains fuse never blew when I ran it through the dim bulb tester, even with all the secondaries connected. The amp works fine through the dim bulb tester and the bulb glows dimly as it should. When I disconnect the 6v and 25v secondary and the standby switch is open the bulb does not light up and the fuse stays intact.

Going straight into the mains, the fuse stayed intact with 6v and 25v secondary unplugged and the standby switch open. When I connected just the 6v secondary, the fuse blew. When I connected only the 25v the fuse also blew. I didn't test it with just the HT because I thought I had ruled it out.

I've been using a 100 watt bulb in the tester. I don't have a 60 watt on hand. I'll have to pick one up.

Can you provide a link to the transformer tester? I can't seem to find it.

[Enzo]

geofex.com

from menu on left, find Tech Tips.  Then within that on the right Tube Amp Tips, and in that is transformer tester.   http://geofex.com/FX_images/xform_test.gif

The bulb should prevent fuses blowing.

Does your meter do AC current?  Pull the mains fuse and insert AC amps meter.  Now power up the transformer with no secondaries connected.  It should draw almost no curent.  But what if the transformer sits there drawing close to an amp?  Then adding some circuit from the board might push it over the top.   This is just one more clue.

[sa230e]

geofex.com

from menu on left, find Tech Tips.  Then within that on the right Tube Amp Tips, and in that is transformer tester.   http://geofex.com/FX_images/xform_test.gif

The bulb should prevent fuses blowing.

Does your meter do AC current?  Pull the mains fuse and insert AC amps meter.  Now power up the transformer with no secondaries connected.  It should draw almost no curent.  But what if the transformer sits there drawing close to an amp?  Then adding some circuit from the board might push it over the top.   This is just one more clue.

Well things are looking pretty bleak for the power transformer.

I desoldered and terminated all the secondary wires, put an ammeter in series with the live wire and ran it through the dim bulb tester and measured 189ma.

I went out and got a 60w light bulb like you said and it's enough current to make the filament glow orange but the bulb doesn't shine. It didn't appear to light up at all with the 100w bulb (with the secondaries disconnected). However the 100w bulb was frosted and the 60w is clear so maybe it did and I couldn't see it. So there's a lesson to be learned, use clear bulbs!

So unless there's a short on the primary side somewhere, I guess my transformer is a giant paperweight now. Testing with my meter, I couldn't find any winding-to-winding or winding-to-chassis shorts but obviously something inside must be busted.

[Enzo]

A winding rarely shorts end to end, all it takes is two turns shorting together, and your meter will never detect that.  That is why RG designed his tester.  Did you try the tester?  It is just a neon bulb and a battery.  Compare a transformer you know is good, to this bad one.

[J M Fahey]

EDIT: Now I saw your schematic, didný know you were using 25L6 , had assumed you were using EL84 or 6V6 or *maybe* 6AQ5, in any case 6.3V filament tubes.
So let´s check the list again:

Going straight into the mains, the fuse stayed intact with 6v and 25v secondary unplugged and the standby switch open. When I connected just the 6v secondary, the fuse blew.

OK, do these tests to differentiate between 2 possibilities:
a) pull all 6V tubes, with transformer secondary disconnected (in fact all secondaries disconnected) measure resistance between 6.3V lines, and from each to ground .
You should measure 400 ohms end to end and 200 ohms from each to ground.
b) connect PT primary to mains, just in case through the bulb limiter, 6.3V secondary to 6.3V filament lines, turn mains on.
Does the fuse blow or bulb limiter shine bright?
If it does, you have a short in that line, somewhere, double check wiring.
If it does not (which I expect) , check you have 6.3V line to line.
If so, plug one 6.3V in the line.
Do you still have filament voltage?
Does fuse blow or bulb shine bright?
If everything normal, repeat with a second tube; and then with the third.
c) pull all 3 6.3V tubes, leave 6.3V secondary connected, and repeat tests with 25V line, measure resistance across it, it should show open.
Set meter to diode test, it might barely show the drop across the bridge rectifier or it might be beyond fullscale, it depends on the particular meter, what we do NOT expect is a plain short (or a coule ohms) across those lines.
If everything fine, connect the 25V secondary.
Do you have 25VAC across 25L6 socket pins? (not tubes in them).
Do you have roughly -35V on C5?
Does bias voltage reach the proper 25L6 grids?
Set it to a high negative bias, meaning cutoff. We are not looking for nice sound yet, just hunting gross shorts.
Plug 25L6 tubes, one by one ... do the filaments light?
If so, plug 6.3V tubes one by one.
As you see, this is an incredibly slow and boring way to reach final working state, but the idea is not to leave sny stone unturned.
Do these tests and report results.
Good luck.

Well things are looking pretty bleak for the power transformer. ??

I desoldered and terminated all the secondary wires, put an ammeter in series with the live wire and ran it through the dim bulb tester and measured 189ma. Mostly magnetization/reactive current so by itself not unusual.

I went out and got a 60w light bulb like you said and it's enough current to make the filament glow orange but the bulb doesn't shine. It didn't appear to light up at all with the 100w bulb (with the secondaries disconnected). so far transformer looks fine.

So unless there's a short on the primary side somewhere, I guess my transformer is a giant paperweight now. why??? It powers filaments (so not open) and does not blow fuses or shine bulb (so not shorted) Testing with my meter, I couldn't find any winding-to-winding or winding-to-chassis shorts but obviously something inside must be busted. so far nothing shows that.

Please run the tests suggested above

[sa230e]

A winding rarely shorts end to end, all it takes is two turns shorting together, and your meter will never detect that.  That is why RG designed his tester.  Did you try the tester?  It is just a neon bulb and a battery.  Compare a transformer you know is good, to this bad one.

No, I didn't try it. I don't have a neon bulb. I'll have to order one.

[sa230e]

EDIT: Now I saw your schematic, didný know you were using 25L6 , had assumed you were using EL84 or 6V6 or *maybe* 6AQ5, in any case 6.3V filament tubes.
So let´s check the list again:

Going straight into the mains, the fuse stayed intact with 6v and 25v secondary unplugged and the standby switch open. When I connected just the 6v secondary, the fuse blew.

OK, do these tests to differentiate between 2 possibilities:
a) pull all 6V tubes, with transformer secondary disconnected (in fact all secondaries disconnected) measure resistance between 6.3V lines, and from each to ground .
You should measure 400 ohms end to end and 200 ohms from each to ground.
b) connect PT primary to mains, just in case through the bulb limiter, 6.3V secondary to 6.3V filament lines, turn mains on.
Does the fuse blow or bulb limiter shine bright?
If it does, you have a short in that line, somewhere, double check wiring.
If it does not (which I expect) , check you have 6.3V line to line.
If so, plug one 6.3V in the line.
Do you still have filament voltage?
Does fuse blow or bulb shine bright?
If everything normal, repeat with a second tube; and then with the third.
c) pull all 3 6.3V tubes, leave 6.3V secondary connected, and repeat tests with 25V line, measure resistance across it, it should show open.
Set meter to diode test, it might barely show the drop across the bridge rectifier or it might be beyond fullscale, it depends on the particular meter, what we do NOT expect is a plain short (or a coule ohms) across those lines.
If everything fine, connect the 25V secondary.
Do you have 25VAC across 25L6 socket pins? (not tubes in them).
Do you have roughly -35V on C5?
Does bias voltage reach the proper 25L6 grids?
Set it to a high negative bias, meaning cutoff. We are not looking for nice sound yet, just hunting gross shorts.
Plug 25L6 tubes, one by one ... do the filaments light?
If so, plug 6.3V tubes one by one.
As you see, this is an incredibly slow and boring way to reach final working state, but the idea is not to leave sny stone unturned.
Do these tests and report results.
Good luck.

Well things are looking pretty bleak for the power transformer. ??

I desoldered and terminated all the secondary wires, put an ammeter in series with the live wire and ran it through the dim bulb tester and measured 189ma. Mostly magnetization/reactive current so by itself not unusual.

I went out and got a 60w light bulb like you said and it's enough current to make the filament glow orange but the bulb doesn't shine. It didn't appear to light up at all with the 100w bulb (with the secondaries disconnected). so far transformer looks fine.

So unless there's a short on the primary side somewhere, I guess my transformer is a giant paperweight now. why??? It powers filaments (so not open) and does not blow fuses or shine bulb (so not shorted) Testing with my meter, I couldn't find any winding-to-winding or winding-to-chassis shorts but obviously something inside must be busted. so far nothing shows that.

Please run the tests suggested above

189ma seems pretty high to me for no load on the secondary.  And that's with tester.  That's a dissipation of almost 23w. Are you sure that's normal?

Anyway, I did tests you prescribed on the 6.3v line.

a) I measured 397 ohms across the 6.3V line.

b) I hooked the 6.3v back up to the transformer winding. The fuse does not blow nor does the bulb shine any more brightly. I added the 12AX7s back one at a time. The tubes light up and the fuse did not blow. The bulb did not appear to shine any brighter - even with all 3 12AX7s plugged in.

One thing I didn't try was to measure the current draw on the primary as I added tubes to see how much it increased. I'm going to have to do that tomorrow before I continue with the rest of the test. I didn't get to the 25v part of the test yet. Hopefully tomorrow I'll have more results for you.

[sa230e]

OK so more results. All results in C are measured with the dim bulb tester with the 60w clear bulb in it.

c) The resistance across the 25v line (disconnected from the transformer) I measured at 5.07MEG. Not exactly open circuit but shouldn't pass enough current to blow the fuse. The diode test showed open circuit.

I measured 25.31v on the filaments of the 25L6s (no tubes).

The grids get negative bias (V4g= -6.15v, V5g=-6.16v)

It's hard to measure the voltage across C5 in circuit as I built the bias supply on a PCB. I cranked the bias on both grids all the way down and measured V4g = -32.3v and V5g=-31.8v so the bias supply looks fine.

V4 and V5 filaments both light up fine.

I also tested the current through the primary as I added each tube back like I said I would. I made a chart out of the results.


I also repeated that test with the 100w bulb as those figures will be closer to what they would be going straight into the wall.


Note that the results with all the tubes in are a bit less accurate as I used a different setting on my meter. The meter has two different probe sockets for current, one is 300mA with a fuse and the other is 10A unfused which has less resolution. I used 10A unfused setting for 4 and 5 tubes as I thought the inrush might blow the meter's fuse but that setting only gives you two decimal places eg. XXX.XXA). On this setting with all the tubes in the inrush current briefly exceeded 400mA so I think that's justified.

[sa230e]

So I re-connected the HT and repeated the tests I did before. So with all the tubes in, 60w bulb in the current limiter and power tubes biased to cutoff the amp pulls about 0.3A on the primary. For some reason, when I biased the tubes (-7.5v on the girds) the current draw was a little less (0.26A). Maybe just a fluke, I don't know. When I repeated the test with a 100w bulb I got 0.3A. Of course I'm using the 10A unfused mode on my meter so I lose some resolution. At this point there is sound from the speaker and I can play through the amp.

I wasn't able get the peak inrush current. By the time my digital meter changes scales the peak is already passed. I do have an analog meter but it can only measure DC current, not AC.