Please refer to Post #3 (by 6L6) of this thread:
https://www.diyaudio.com/community/threads/f6-illustrated-build-guide.277850/#post-4406647
R11, R12 are gate stopper resistors, and their value can be anywhere between 47 Ohms - 680 Ohms. Looks like 6L6 used 110R. The gate stopper resistors are there to prevent ultrasonic oscillations.
What value do you have in place for R11, R12?
Bruce,
What ZM said is that you can test the mosfets, though not on the PCB. zman01 has already linked something for testing with a multimeter. The other testing ZM mentioned is a matching type circuit, for example: https://firstwatt.com/pdf/art_matching.pdf
Basically you can use the Fig 12 (N-channel) circuit to measure the Vgs at some known current (the exact value not important) to see if the mosfet is working.
Your bulb tester remained lit because your amp was drawing a fair amount of current, mostly likely from the right channel. Assuming there isn't a short somewhere and the mosfets are ok, one probable reason is that P1 and/or P2 were set 'high' with high Vgs voltages going to the mosfets (and causing the mosfets to draw a lot of current).
If you do not yet feel comfortable removing and testing (or replacing outright) the mosfets, perhaps you can first try the following:
1) With power off, remove the left channel from power supply, leaving right channel attached to power supply.
2) Turn down both P1 and P2 on the right channel (to lower the Vgs to the mosfets)
3) Power up with bulb tester in place. If 2) went well and the mosfets are OK the bulb tester should brighten and then dim.
4) Measure the voltages across the gate and source pins of the mosfets. They should be low. (The mosfets don't turn on until you see a few volts here.)
5) Confirm the rail voltages and measure the the voltages across the sets of three LEDs. Are the LEDs on? If so, do you see 5 to 6 volts? If the LEDs are off, do you see something significantly higher (say, around 14V)?
If the LEDs are off and you see a high voltage, power down and turn the LEDs around. Power up again. Hopefully the bulb tester still passes and then check if the LEDs are now on and you have 5 to 6 volts across them. Verify that the voltages across the gate and source pins of the mosfets remain low. You can turn up the pots a little bit and check that you see a voltage change across the gate/source pins.
If all this checks out then you can power down, remove the bulb tester, and try the biasing procedure again.
What ZM said is that you can test the mosfets, though not on the PCB. zman01 has already linked something for testing with a multimeter. The other testing ZM mentioned is a matching type circuit, for example: https://firstwatt.com/pdf/art_matching.pdf
Basically you can use the Fig 12 (N-channel) circuit to measure the Vgs at some known current (the exact value not important) to see if the mosfet is working.
Your bulb tester remained lit because your amp was drawing a fair amount of current, mostly likely from the right channel. Assuming there isn't a short somewhere and the mosfets are ok, one probable reason is that P1 and/or P2 were set 'high' with high Vgs voltages going to the mosfets (and causing the mosfets to draw a lot of current).
If you do not yet feel comfortable removing and testing (or replacing outright) the mosfets, perhaps you can first try the following:
1) With power off, remove the left channel from power supply, leaving right channel attached to power supply.
2) Turn down both P1 and P2 on the right channel (to lower the Vgs to the mosfets)
3) Power up with bulb tester in place. If 2) went well and the mosfets are OK the bulb tester should brighten and then dim.
4) Measure the voltages across the gate and source pins of the mosfets. They should be low. (The mosfets don't turn on until you see a few volts here.)
5) Confirm the rail voltages and measure the the voltages across the sets of three LEDs. Are the LEDs on? If so, do you see 5 to 6 volts? If the LEDs are off, do you see something significantly higher (say, around 14V)?
If the LEDs are off and you see a high voltage, power down and turn the LEDs around. Power up again. Hopefully the bulb tester still passes and then check if the LEDs are now on and you have 5 to 6 volts across them. Verify that the voltages across the gate and source pins of the mosfets remain low. You can turn up the pots a little bit and check that you see a voltage change across the gate/source pins.
If all this checks out then you can power down, remove the bulb tester, and try the biasing procedure again.
Hello everyone, I apologize in advance as I'm sure the question must have been asked already (surely many times). I'm afraid I can't find the answer through my search on this thread...
What are acceptable AC measurements on each rail of the PSU? Getting a little less than 40mV on both (no load connected).
I have a 400VA transformer and followed all values on 6L6's BOM.
Cheers!
What are acceptable AC measurements on each rail of the PSU? Getting a little less than 40mV on both (no load connected).
I have a 400VA transformer and followed all values on 6L6's BOM.
Cheers!
Sorry, I was being a little lazy, I should have thought to look this up on Mouser, where I bought the mosfet.
I pulled the two Mosfets last night, which may I say was a bitch of a job, getting three pins clean that close to the board, even with a desoldering tool and solder wick. In the end I found that by holding them with a handkerchief to protect my fingers from the heat I could melt and wiggle, pin by pin leaving a weakened solder joint that eventually gave way. I have not tested them yet but I ordered some new ones today since I suspect if they were not damaged to begin with they almost certainly are now!
Anyway, powered up the board, got the leds all to light and measured voltage across the gate and source of Q1 and Q2, both could be made to range from 0-5.8v by adjusting the pots, this was on the dim bulb tester so I expect voltages will increase off the tester. Voltage across the 3 LEDs 5.8. Things are looking good. Will test the Mosfets later but the undersides are a bit bubbled which does not bode well
See pics.Once again muchas gratias for your guidance.
Attachments
Getting new mosfets is the right call. Anything showing thermal damage like that should not be used.
Happy that you can get 0 to 5.8V adjustment with the pots. I suggest you leave them in the 0V
positions before you install new mosfets.
And before you start on your left channel mod, also please turn down P1 and P2 so that the mosfets are off
so you don't have to worry them after the mod.
Have fun.
Happy that you can get 0 to 5.8V adjustment with the pots. I suggest you leave them in the 0V
positions before you install new mosfets.
And before you start on your left channel mod, also please turn down P1 and P2 so that the mosfets are off
so you don't have to worry them after the mod.
Have fun.
sndlvr,
I had experienced blown MOSFETs (when trying a mod) a few days back on one channel, and then reverted to turning down P1 and P2 by powering up the amp without the MOSFETS connected. Followed the same procedure that Dennis Hui has suggested to you, and now the channel is up and running with new MOSFETs installed.
I had experienced blown MOSFETs (when trying a mod) a few days back on one channel, and then reverted to turning down P1 and P2 by powering up the amp without the MOSFETS connected. Followed the same procedure that Dennis Hui has suggested to you, and now the channel is up and running with new MOSFETs installed.
If anyone has been tracking my build, just up'ed the game on the PSU. Put in an Antek 500VA transformer and the diyAudio PSU board rectifier section in prep for monsterblocks. Real estate is getting tight.
Interestingly, the PSU output sits right at +/-24V as opposed to the PSU it replaced (Hammond 220VA transformer w/ discrete bridges) that sat at +/-26V. The bias on the transistors has shifted from 600 mV down to about 580 mV. Did not touch the pots in the install process. Offset is as much 0 as my Fluke will measure. Will rebias tomorrow morning after I let it sit some more.
Getting the right channel PSU swapped out tomorrow, I hope. Maybe next weekend will add the second set of boards to attempt balanced "monsterblocks" Will see what the offset is like between the + and - "channels" and decide the input/output configuration on the rear panel.
Interestingly, the PSU output sits right at +/-24V as opposed to the PSU it replaced (Hammond 220VA transformer w/ discrete bridges) that sat at +/-26V. The bias on the transistors has shifted from 600 mV down to about 580 mV. Did not touch the pots in the install process. Offset is as much 0 as my Fluke will measure. Will rebias tomorrow morning after I let it sit some more.
Getting the right channel PSU swapped out tomorrow, I hope. Maybe next weekend will add the second set of boards to attempt balanced "monsterblocks" Will see what the offset is like between the + and - "channels" and decide the input/output configuration on the rear panel.
Attachments
Last edited:
I desolder the legs until I can get one to move (wiggle the mosfet and watch the legs. I then load up the other two legs with solder, heat both legs with soldering gun at the same time and wiggle the mosfet out. Still a pain, but the best way I know of so far. Holding the mosfets with your hands makes it so you don't overheat the mosfet. Although they are very tough. I use denatured alcohol to cool it down if needed to.
Best of luck!
What you had in #3729 were monolithic bridges. What you have in #4137 are bridges with discrete diodes.
A couple potential reasons for the voltage change are different winding ratios between the transformers and/or a higher voltage drop through rectification.
Seems like a tweak of the bias, and you're ready to keep the covers on. Thanks for posting all the progress and questions. What a fantastic reward for sticktoitiveness. Two fantastic amps. Congrats!
.
.
.
.
So, whatcha gonna build next?
