Crack+Speedball Voltage issues + smoke [SOLVED]

[Codeine]

I got the kits few days ago and thought to do crack+speedball together. I finished all the soldering and put the tubes in. Turned on the crack and waited few seconds. They started glowing at the bottom. As soon as i went to grab my voltmeter to start the voltage check, I heard a pop and a spark on the left side. Immediately I turned it off. Visual inspection didn't reveal any damaged parts. So I turned it on again and did the voltage check. Everything checked fine except T1 and T7l both were around 160 V. I also noticed the LEDs on one of the small boards didn't light up properly. the one close to the MJE350 looked off and and the one next to was very dim. All other LEDs were bright. Turned it off, and tried to troubleshoot more. I reheated the parts on the small board and nothing changed. I did another voltage check but this time i noticed after say 30 seconds the LEDs began to fade. I was checking T9 and the voltage dropped from 100 to 30 V. Then I saw smoke coming out from the rear, I looked over and saw the long 270 ohm resistor smoking and glowing kinda red in the middle. I turned the amp off and have stopped proceeding. I checked for loose connections and all the joints looked firm with enough solder on them. What should I do now?
Thanks.

 

[Paul Birkeland]

The first thing to do is to remove the Speedball and build the Crack stock, with the resistors.  This is absolutely mandatory before proceeding with the rest of the installation.

From what you have mentioned, I'd imagine that the TIP50 transistors aren't installed properly, but it's not even worth looking at that until you know that you have a stock, functional Crack.

-PB

 

[Codeine]

Thanks for the advice PB. I took out the Speedball and soldered back the 3K and plate resistors. Voltages all check out. A1, A2, B1, B4 are 80 V instead of 90. I guess those are within the limits. So the problem is from the Speedball. How can I make sure the TIP50 transistors are the problem?

 

[Paul Birkeland]

Did you install the mounting hardware as specified in the manual?  Recently, we have seen a reasonable number of folks leaving off the mica insulators or not using the shoulder washers.  Leaving out either part will cause the issues you describe.

Also, you can remove the 22K resistors and install the small boards, then run the amp.

The smoking resistor(s) in the power supply almost always come from an issue with the 6080, as the 12AU7 can't draw enough current to cause that kind of abuse. 

-PB

 

[Codeine]

Did you install the mounting hardware as specified in the manual?  Recently, we have seen a reasonable number of folks leaving off the mica insulators or not using the shoulder washers.  Leaving out either part will cause the issues you describe.

I exactly assembled them the way the manual said.

Also, you can remove the 22K resistors and install the small boards, then run the amp.

you mean just replace the 22k resistors with the small boards? and not install the big board, right? how about the cathode resistors? what voltages should I expect?

Thanks.

 

[Paul Birkeland]

Yes, the small boards can be installed in place of the 22K resistors, as indicated in the manual, and the 3K resistors can stay.  The amp will work fine this way, and you can measure the voltages at T1 and T5 to check your work on the small boards.  This will also let you listen to the amp if you pass the voltage checks.

 

[Codeine]

I installed the small boards. One of them does not work. Only one of the LEDs lights up (very dim). T1 stays around 150 V.

 

[Paul Birkeland]

Try swapping small boards, side to side, to ensure that the board is defective.

 

[Codeine]

Yup I did that too. The defective board doesn't work on the other side (the other one works on either side and gives correct voltages at its terminals).

 

[Codeine]

quick update: I checked the diodes on the defective boards with the multimeter on diode mode. D1 wouldnt pass so I looked closer and found out that one of the leads was broken off. So i replaced the diode with one of the diodes from the big board. Now both small boards work and voltages settle to correct values. I just need to get a new diode. Any suggestion as to where to get a replacement diode?

 

[Paul Birkeland]

The diode needs to be an HLMP-6000.  These are available from Mouser online, and other electronics retailers.

Still, a broken diode on one of the small boards wouldn't cause the power supply to overheat, so I suspect there to be an issue on the large PC board still.

-PB

 

[Codeine]

Thanks. I suspect that too. I'll be ordering some extra TIP50s just in case. I also noticed the 270ohm wirewound resistor that goes across the terminal strips gets extremely hot (the smoke was coming from this resistor). Can it be replaced by another resistor with higher power rating?

 

[Paul Birkeland]

This resistor will get overly hot when there is a fault in the circuit, which would eventually blow the fuse.

That resistor and the other 270 Ohm resistor handle the same amount of current, but the one that goes across the power transformer is easier to notice.  In normal operation, that 5 Watt resistor sees about 1.3 Watts of power, which leaves a very large safety margin, and the operating temperature in that situation is well below any threshold of concern.

You can replace that resistor with a 10 Watt part, but the real solution is to fix the issue cooking it in the first place.

-PB

 

[Codeine]

Well I triple checked everything. All the connections and joints seem to be in good shape (even reheated most of them). The voltage across the long 270 Ohm resistor is around 20 V, so it's dissipating around 1.5 Watts if i'm not mistaken. Also the 3k resistors get very hot too. I checked the bias points on the small boards they all make sense. I even reverted back to stock crack and still these resistors get very hot. I'm not sure what to make of this. I ordered the extra diodes today, so it'll take few days before I can proceed with the big board. As of now, I hear no noise or distortion in the sound. Maybe the tubes play a role in this? Can the 12AU7 cause this kind of a problem? the getter in my 12AU7 is a little tilted, is that normal? does it affect anything?

 

[Paul Birkeland]

The resistors are supposed to get hot.  You've verified their dissipation by measuring the voltages.  If you'd like, you can hit them with an IR thermometer to look at their actual temperatures.  I believe the big resistors are rated to run at 150C, but we run them much cooler.  The tubes will also get hot, out of necessity, and this is nothing to be alarmed by.

As long as the flashing is silver and not white in the 12AU7, the getter can do its job.  The little metal ring or plate that sits under the flashing is not used once the flashing is drawn onto the glass. 

-PB

 

[Codeine]

I received the parts today. I soldered in the missing LED on the big board. Before installing the big board I checked the mounting of the TIP50s. To my surprise, the metal contact on one of the TIP50s was shorted to the heatsink and therefore shorted to ground! I desoldered it and took it apart. The plastic lock washer was melted around its shoulder. I straightened it so it could hold the screw without letting it come in contact with the heatsink. I threw in a new TIP50 and put them all together but without the flat washer to let the lock washer's shoulder completely secure the screw. I hooked it up and the voltages all checked out. This might explain the spark on the first run. The 270 Ohm resistors still get very hot and the heatsinks too. Any ideas on how to make the mounting more reliable around that part? Because the washer might melt again after long use and short the metal side to ground.

The amp runs great with my optical Modi. No hissing or noise so far

Edit: I also got a 10W 270 Ohm wirewound resistor to replace the long 5W resistor later.

 

[Paul Birkeland]

The shoulder washer on the TIP50 would melt from the short, not from general operation.  We can pop a replacement in the mail if you'd like, as that part is quite important. 

 

[Codeine]

Ok, so I did some modification to the TIP50 mounting, and I think it'll work fine. So I replaced the screw with a Nylon screw and a nylon flat washer in the back with a nut. Also replaced the 270 ohm resistors. Put in a 10W as the long one and a 7W as the other one. They run cooler now I think (at least the 10W resistor). Also did a temp measurement:

10W 270 Ohm (longer one) = 55 C
7W 270 Ohm = 95 C
6080 Tube = 104 C
12AU7 Tube = 60 C

 

[Paul Birkeland]

I would be a little concerned about the long leads on that 10W resistor.

Also, were you able to get enough tension on the nylon screw?  If the TIP-50 isn't tight against the heatsink, it will overheat. 

-PB

 

[Codeine]

I'm planning to cover the leads with some heat-shrink wire covers. I used an 8-32 screw which turned out to be a tad bit wide for the heatsink hole, so I had to "screw" it in with a little force. Overall, with no room to play and the nut pulling it on the other side, it pushes the transistor very tightly against the heatsink.