• Disclaimer: This Vendor's Forum is a paid-for commercial area. Unlike the rest of diyAudio, the Vendor has complete control of what may or may not be posted in this forum. If you wish to discuss technical matters outside the bounds of what is permitted by the Vendor, please use the non-commercial areas of diyAudio to do so.

Tubelab SE finished but need help.

I'm stuck after doing the first round of test. After I finished connecting all the transformers, I did the "Checkout" process. The voltages checked out fine and went as far as playing some music. Then the fuse blew after about 10 minutes. I disconnected speaker, music source IC and pulled out all the tubes. Replaced the fuse and redid the "Checkout" process. At the step where checking the voltage of output tube grid, both sides measure negative voltage but after about 10 seconds of powering up, one side starts to show positive voltage which wasn't the case during my first round of test. What could be doing this?
 
Sorry for the slow reply, I was out of town yesterday, and at the warehouse most of today. I will be out until 9PM EST tomorrow, but I will answer any additional questions when I get home.

one side starts to show positive voltage which wasn't the case during my first round of test. What could be doing this?

This is usually caused by a bad mosfet in the channel that has the problem but it could be a bad resistor or the bias pot itself could be bad. Unfortunately the easiest way to tell if a mosfet is bad is to remove it from the PC board. If the voltage at the gate terminal behaves normally (adjusts from - 70 or so to -10 or so) without the mosfet installed, the mosfet was bad.

What type of mosfets did you use? The mosfets that are recommended replacements for the 2SK2700 seem to be more sensitive to heat than the good old 2SK2700. Make sure the mosfet is securely attached to the heat sink.

How much B+ voltage are you getting (measured across R30) and what is the B- voltage (connect the meter across R7). Some transformers seem to cause a higher B- voltage than others making the mosfets get too hot. I have an amplifier sitting here with a 272JX in it that has played fine for the last 3 weeks.

While I'm at it, power transformer (Hammond 272JX) buzzes and I can feel the vibration. Is it normal? I heard that it's common for Hammond PT.

I have heard this also, but mine makes absolutely no noise even with my ear next to it.

Also, there are 9 leads off of it but the schematic shows 8 are connected. Do I leave one floating?

The yellow with black stripe lead is not used and should be taped up or cut off. This wire has the full B+ voltage on it so it must not touch anything. The green with a yellow stripe lead is not used with a 300B amp although it won't hurt anything to connect it to the PC board. Pad 1 is not connected with 300B tubes.
 
tubelab.com said:
What type of mosfets did you use? The mosfets that are recommended replacements for the 2SK2700 seem to be more sensitive to heat than the good old 2SK2700. Make sure the mosfet is securely attached to the heat sink.
I used Fairchild MOSFET. I ordered an extra one. Should I try without it or go ahead and swap?

How much B+ voltage are you getting (measured across R30) and what is the B- voltage (connect the meter across R7). Some transformers seem to cause a higher B- voltage than others making the mosfets get too hot. I have an amplifier sitting here with a 272JX in it that has played fine for the last 3 weeks.
Without any tubes, B+ is -21 vdc, B- is 242 vdc. With rectifier tube in, B+ is 432 vdc, B- is 184 vdc.
 
In addition, when I was adjusting output tube filament current of now defective one, I once turned the pot little past what's ideal for 2A3 but then turned it back down. Would that have any affect on MOSFET? Did I do it right when measuring the current, I set the multimeter to "200m" (which is next to "10A" on dial), then adjust the pot to see 2.5 on the display?
 
tubelab.com said:
This is usually caused by a bad mosfet in the channel that has the problem but it could be a bad resistor or the bias pot itself could be bad.
It was MOSFET. That fixed it however, I've made a rookie mistake. While testing input tube (5842) plate to ground voltage, right channel ground lead clip loosened and disconnected. Instead of powering off first, I reconnected it right away. Then I thought I heard a little spark at the tip of lead and something near the right input tube started to buzz. Then the tube started to glow brighter than the left one. It smelled like something was burning. Did I fry semiconductor U2? :hot:
 
If the 5842 is overheating the CCS chip U2 is most likely at fault. This may also cause R10 and R16 to overheat. Measure the voltage at the plate of the overheating 5842 it should be near 175 volts and adjustable by R9 if the voltage is too high and not adjustable the chip is bad. Do not operate the amplifier for more than a few seconds at a time or the tube may be damaged.

R14 is in a different circuit than the 5842 - U2 circuit. It sets the current through the mosfet. If R14 is severely overheating there is still trouble with the mosfet. Is the output tube grid negative and adjustable by R12? If so and the resistor is still overheating, is the resistor the correct value?
 
Update

Replaced mosfet and solved the problem with R14. Now for 5842 plate to ground voltage check, both sides are adjusted to 175 vdc but right side (earlier problem) tube is glowing brighter in about a minute after the power up. I switched the tubes but the right side still glows brighter. Could it be something to do with R10 and 16?
 
It is possible that the U2 chip could be damaged in a way that allows too much current to pass. I have only seen this happen, but it was in my amp and it happened after I was poking around in a live amplifier with a scope probe. Measure the voltage across R16. It should be about 3 volts to give 10mA. If it is significantly higher, the chip is toast.
 
I replaced the damaged U2 chip. Once the rectifier tube warmed up, I saw a spark at the base of leads where the chip is soldered to the board. What could be doing this now?

If you saw a spark there is either something causing a short on the board, like a small solder fleck, or there is a cracked trace on the board, and you saw an arc jumping the gap.

I am at a loss to figure this out without seeing the board. Please send me an email.
 
I MIGHT have pulled a pad off my tubelab pcb. It was for an unused pin on one of the tubes, so no big deal. I was also doing some not so standard soldering of little copper tubes to the pads. On the other hand, I might have not heated the pad up enough, so it was just a cold solder joint that came loose. I say MIGHT, because the solder came up perfectly flat as if the pad released from the pcb, but the pad on the PCB was still gray, tho not shiny. I didn't do any further testing because the pad isn't used.

Anyway... I've pulled pads off of plenty of small home made PCB's in the past from overheating, and that was with a lower temp iron.

Point being.... check the continuity from the legs of those chips to whatever it is they connect to next on the PCB.