Finished Build - Normal Resistance/Voltage - No A8 LED Light - Loud Noise

[Tommy Coyote]

Hi All,

I'm new to soldering and tube amps in general, and had a lot of fun with my Bottlehead Crack build, but could use some help as something didn't go right. When performing the resistance checks everything came out normal. The glow test looks good as well (both halves of both tubes appear to light up fine). The voltage checks also look good to me, but I will post all measurements here just in case that helps.

The A8 LED lights up initially, then flickers a little, then goes out within a couple seconds. This behavior seems to occur after each cooldown and test. The A3 LED takes a few seconds to slowly light up, but then it stays lit. I went ahead and tested the sound with my headphones (on my initial inspection I THOUGHT the A8 LED was lit, but now realize it was just reflecting light from elsewhere)... with nothing plugged into the RCA jacks, there is initially no sound, then a very loud buzz that persists regardless of the volume knob setting. Adding an input source makes no difference, but picking up the unit and turning it on/off the sound seems to follow the A3 light (a popping sound synced to a flicker as it goes out... the loud noise kicks in when the light gets bright).

I searched the forums as well and found a post with similar symptoms so I measured the voltage at A3 and A8 and included those below. I tried wiring a cable from A3 to A8 as well but that had no effect on the voltage measurements (and I checked with headphones again... same loud sound).

T1: 84.9V (slowly dropping to 77.5V)
T2: 169.3V
T3: 0.4mV
T4: 169.3V
T5: 85.2V
T6: 0.4mV
T7: 121V
T8: 0.7mV
T9: 103.7V
T10: 0.4mV
A3: 1.567V
A8: -12.7mV

I went ahead and ordered a fresh pack of HLMP-6000 LEDs and plan on replacing the A8 LED as soon as possible. Any other suggestions are much appreciated

Thanks!
-Tommy

[Paul Birkeland]

T1: 84.9V (slowly dropping to 77.5V)
T2: 169.3V
T4: 169.3V
T5: 85.2V
A3: 1.567V
A8: -12.7mV

This set of voltages is a little improbable.  If you have -12mV at A8, terminal 1 should have also have a voltage issue.

I would reheat the joints at 1U, 1L, 2U, and 2L.  B4 also looks like it could use a little extra solder.  All of the lower terminals where the UF4007 diodes are mounted need more heat/solder.

When an LED isn't working, it's really important not to get fixated on the LED itself, as there are many causes for the LED to not illuminate. 

[Tommy Coyote]

Thank you so much for the speedy reply!

I reheated and added a little more solder to 1U, 1L, 2U, 2L, B4, 18L, 19L, 20L, and 21L. All of the resistance checks were as expected, and the voltages and LED behavior was the same (I didn't dare connect headphones this time). I lifted the unit and confirmed that both halves of both tubes are lightning up as well.

Updated voltage measurements and photos are below. I took a second (and longer) reading of terminal 1 at the end and confirmed it is holding steady at 76.8V, but A8 is still really low. The A8 LED slowly lit up with A3, then blinked out after a couple seconds (A3 remains lit).

So is the 12AU7 tube getting voltage on A6 from terminal 1, and is that meant to pass through the tube to A8? If the tube is lightning up on both sides (can clearly see two "dots" on top), is the tube likely fine? And likewise with the A8 LED... if it starts to light up initially then blinks out quickly, it's probably not the LED as you mentioned, correct?

I'm really lost as to what to try next, and I appreciate your assistance!

T1: 83.5V (slowly dropping to 76.8V)
T2: 167.5V
T3: 0.5mV
T4: 167.7V
T5: 81.9V
T6: 0.4mV
T7: 120.2V
T8: 0.6mV
T9: 101.8V
T10: 0.3mV
A3: 1.577V
A8: -13.4mV

[Paul Birkeland]

Can you very carefully measure the DC voltage at A7?

As the tube conducts, current flows up through the LED to the tube cathode, then to the plate.  You have a voltage drop across each 22K resistor, and this indicates that you have some current flowing through both halves of the tube.

The negative voltage and one dark LED doesn't make a lot of sense though.  With negative voltage at A8, there should be no current flow.

It is remotely possible that the 12AU7 experienced some shipping damage, though this is incredibly rare. 

[Tommy Coyote]

Just fired it up and measured A7: it's at 0.3mV with the volume knob turned all the way down, and at -0.743V with the volume knob turned all the way up.

[Paul Birkeland]

Just fired it up and measured A7: it's at 0.3mV with the volume knob turned all the way down, and at -0.743V with the volume knob turned all the way up.

Well, that's not right.  How are the solder joints on the volume pot where the black wires attach? 

It may be worth trying a different 12AU7 at this point too.

[Tommy Coyote]

I took a few photos of the 12AU7 in case that helps, unplugged the tubes, and checked the joints as suggested. They looked OK to my novice eyes but I took a resistance measurement from the solder tabs on the RCA jacks to the center (plate?) of the 9-pin socket and it was at 0. I reflowed the two inner joints of the volume pot (upper and lower), then re-measured and got 0 ohms again.

I plugged the tubes back in, turned it on and carefully re-measured the DC volts for T1-5, A3, A7, A8, and got not change there. Before and after photos of the volume pot solder joints are attached ("after" I reflowed them is the one without the flash, and with a little singe on the white wire hehe).

Is it best to obtain a replacement 12AU7 from Bottlehead, or would one of the ones I see on Amazon or other online shops suffice? Any brand recommendations?

Thanks again for your help! I'm happy to try more fixes or troubleshooting steps of course, just not sure what else to try.

[Paul Birkeland]

You can e-mail replacementparts(At)bottlehead(dot)com and request a replacement.  Please include a link to this thread and let them know that this is my recommendation.  Also don't forget to provide your name and shipping address, and if possible your original order number. 

I would also put the jumper back between A3 and A8, then recheck the voltage at A7 with the volume pot all the way up.  Can you also let me know what the resistance is between each outside pot lug and ground?  How about each middle pot lug and ground? 

At this point I think it's 50/50 that you either have a pot that needs some help or a tube that needs some help.

[Tommy Coyote]

At this point I think it's 50/50 that you either have a pot that needs some help or a tube that needs some help.

And 100% that I need some help lol.

Here are the resistance measurements:
Outside Upper Pot: 94.9kΩ
Outside Lower Pot: 94.3kΩ
Middle Upper Pot: 12.62kΩ
Middle Lower Pot: 13.22kΩ
And this was obvious but the inner pots are both at 0Ω.

I wired A3 to A8 and the DC volts for T1-5 were still in range, but here's the new voltage measurements with the volume turned up:
A3: -28.3mV (light on)
A7: -0.742V
A8: -13.5mV (both lights off)
A3 again: -13.5mV (both lights off)

The A8 LED never lit up. A3 was on for most of the time, but while taking measurements it turned and stayed off (hence the 2 measurements). I poked around a little with a chopstick and tried wiggling the 12AU7 a bit but there was no flicker.

Emailing for a replacement 12AU7 now, thanks!
-Tommy

[Paul Birkeland]

but while taking measurements it turned and stayed off (hence the 2 measurements).

There's a flaky solder joint in your amp.  This is definitely a symptom of a joint that isn't well connected.  Also with the LED on and A3 wired to A8, your DC voltage should be 1.5V.  There isn't any way that A3 and A8 can have different voltages if they are wired together.

I am now less convinced that you have a tube problem and more convinced that there's still a flaky solder joint.

[Tommy Coyote]

There isn't any way that A3 and A8 can have different voltages if they are wired together.

To clarify: A3 and A8 both measured -13.5mV. A3 was -28.3mV initially, then the light went off before I got to A8, then both were -13.5mV. Sorry for the confusion.

Any suggestions on where to start resoldering? The voltages all measure within range outside of the A socket, and the resistance is all normal (I think). It's easy to believe I didn't do a perfect soldering job since this is my first project of this type, but short of pulling everything out and starting over I'm not sure where to focus.

Thanks!

[Deluk]

The led goes from the centre black to A3. There is no other wire on A3 (on my Crack) the red wire links across to A4 from A5. That's not what I'm seeing in the last picture.

[Paul Birkeland]

From your photo, A9 looks not all the way soldered as well.

[Deke609]

The led goes from the centre black to A3. There is no other wire on A3 (on my Crack) the red wire links across to A4 from A5. That's not what I'm seeing in the last picture.

It looks like the extra red wire is a jumper connecting A3 and A8.

From your photo, A9 looks not all the way soldered as well.

If it were my amp I'd redo at least A9 (green wire) and the center pin. I'd desolder both and then redo them by firmly crimping the wire lead to the terminal lugs before resoldering.  You'll need either a solder sucker or desolder braid. If you use braid, make sure you get the kind with no clean flux - and don't even try the kind without flux b/c it won't do anything except overheat your components.  I'm a big fan of MG Chemicals' 2.5 mm "No Clean Super Wick". For desoldering the center pin, I'd try to apply the heat as far away as possible from the LEDs to minimize the chances of frying them. My bet would be the low end of the center pin hole.

cheers, Derek

[Tommy Coyote]

Awesome, thanks for all the feedback everyone! I plan on redoing A3, A8, A9 and the center pin of the A socket. I may rotate the two LEDs just to see what happens (since I'm pulling them anyway).

...the red wire links across to A4 from A5.

I've seen photos of Crack kits with red wires from A4 to A5, but my version uses a green wire, which you can't really see in this photo but looks right to me. I appreciate the callout though 

It looks like the extra red wire is a jumper connecting A3 and A8.

Correct, although I'll be trying again without that extra wire.

If you use braid, make sure you get the kind with no clean flux - and don't even try the kind without flux b/c it won't do anything except overheat your components.

Ahh this explains why the stuff I experimented with earlier didn't work as well as what I've seen in videos.

I'll post updated results soon, thanks!