Bottlehead Crack - faulty valve

[wmccann]

Have an issue with my recently built crack that is driving me crazy.

Was "running the amp in" with pair of HD660S2's and had it running for 10 days - 8 hours on, 4 hours off, music streaming.

All good, then while It was playing away suddenly there was a crack and the volume went up to max on its own - As I went over to the headphones the volume reduce and all seemed good.

This is when the trouble started, essentially the left channel was making a lot of noise, was either full volume or noisy etc.  I traced this down to the cenre and right pins on the potentiometer were bad - they were not making good contact with the actual resistor strip.  I tried to crmp them harder but it would work for about a minute when playing music and then stop working.

I also noticed that the front LED (left channel?) on the 12AU7 was not working.  tested the voltage across the led and when I touched the multimeter across the led bingo it lit up and worked for about 30 mins then went out again.  Now it does not work at all.

I traced the voltages on the crack and on the small board the 0A output is 60v but the 0B output is between 35 and 40v.  I have tried everything but cannot get it to work correctly.

Steps so far:

1) tried swapping the output from the crack (0A and 0B) - voltage reduction moved
2) disconnected all upstream items from the 12au7 - no difference
3) disconnected the Crack large board 200v power - no difference

At this point the only thing I can think of is the tube is gone - could a faulty potentiometer cause the tube to blow?  (or vice versa?)

Also, I noticed that the 6.3mm Jack the top pin does not seperate correctly due to the "nipple" most jacks have - seems that sometimes it can short with the resistor - is there a better jack that one can get (I can't find one so far unfortunately and I assume the resistors are there to protect the tubes if it is switched on with no headphones attached.

Any help appreciated,

W.

 

[Paul Birkeland]

I think it would be a good idea to slow down a little bit.  When you have an issue with the amp, the first step absolutely is a set of voltage checks. 

The volume going up to maximum would be the volume pot losing its ground connection, which is most often a soldering issue.  If the lugs have been squeezed on where they mount to the pot wafer, then replacement will be in order.

The operation or lack of operation of the LEDs on the socket depends a whole lot on what the actual DC voltages are, and resoldering/poking at the LEDs because they don't light up is usually only going to damage the LED itself. 

I have never, ever in my 25 years of doing this seen a tube destroy a volume pot in the manner you describe.  That doesn't mean it couldn't happen, but this is incredibly unlikely.

On the headphone jack, the tip will not push the spring contact up all the way, but if you are getting sound out of the left channel, it's pushed up enough.

 

[wmccann]

Thanks Paul,

I did a full voltage check and the only voltage that was out is A1 - 40v instead of 60v.  As said if I swap A0 and A1 around A0 reads 40v and A1 reads 60v.  Sound on the left channel is very distorted.

I did check the voltage across the LED (as per another post) and for the left channel it is 0.55v vs >1.5v for the right channel.

The volume potentiometer is definetly bad - the lugs are "crimped" onto the board - middle lug is deffo not making proper contact and the right hand one is not great.  I am wondering if this could have caused the tube to go (and not vice versa)?  note the strip is giving a resistance (over left/right lugs) of approx 40k rather than 100k (which the right channel gives)

Have already ordered a different tube and an alps volume control as the control was very noisy from the start - so will test them and see if that resolves the issue. 

Is there any way to test the tube (apart from wiring it up in reverse - left side to right side and right side to left side and seeing if the "right side" exhibits similar behaviour)?

re the headphone jack - agree but what I found was I had to twiddle the jack to get sound in the left channel - I think it was not opening enough - is it right to assume that if it is closed with the headpones connected and with the resistor in parallel it won't overstress the amp?. 

W.

 

[Paul Birkeland]

I did a full voltage check and the only voltage that was out is A1 - 40v instead of 60v.

Can you post a build photo of your front Speedball board?

The volume potentiometer is definetly bad - the lugs are "crimped" onto the board - middle lug is deffo not making proper contact and the right hand one is not great.  I am wondering if this could have caused the tube to go (and not vice versa)?

This will just cause the amp not to work properly, so a  replacement is in order.

Have already ordered a different tube and an alps volume control as the control was very noisy from the start - so will test them and see if that resolves the issue. 

It's important to note that you can't solder wires directly to the pins on the Alps pot, as those pins will get overstressed by direct wiring and are likely to snap off.  Be sure to get one of the companion PC boards to prevent this issue.

Is there any way to test the tube (apart from wiring it up in reverse - left side to right side and right side to left side and seeing if the "right side" exhibits similar behaviour)?

On the section of the pot that's not working, move the middle wire to the other level of the pot to put the amp into mono mode.  This will send the working channel to both halves of the 12AU7, then you can do a voltage check and evaluate the functionality of everything else.

re the headphone jack - agree but what I found was I had to twiddle the jack to get sound in the left channel - I think it was not opening enough - is it right to assume that if it is closed with the headpones connected and with the resistor in parallel it won't overstress the amp?. 

If you have to wiggle things, it's possible that one of the lugs on the headphone jack was overheated and melted the plastic enough to move.  With no headphones plugged in, the 2.49K resistors are shorted out. With the headphones plugged in, the 2.49K resistors are in parallel with the headphones.

 

[wmccann]

Ok, Think I have found one of the problems, the LED was bust (I reflowed the solder - it must not of liked it)- it fell apart on me when I tried to test it.  Have replaced (though not with a HLMP 6000 but a micro red LED I had lying around - (will this make a difference?)

Will test voltages tomorrow.

I don't believe I overheated the headphone socket but the 6.3mm jack I was using looked like the Taj Mahal - the groove was very big - anyway have sorted this now so that probelm is resolved.  Both channels are now working over the bottom of the pot (mono mode).

Thanks for the tip re the ALPS, will make up a breadboard.

BTW, tested the 12AU7 using some cables I had and both sides working fine - LED was the issue.

Thanks again for your help,


W.

 

[Paul Birkeland]

The LED we select is very quiet and provides a very stable voltage reference.  You'd want to make sure the red LEDs you're using provide about 1.5V of bias.  If they are up closer to 2V of bias, that can cause some issues downstream of the 12AU7.