Something doesn't seem right. 2 6DJ8s should not be pulling 1.8A. That's enough current to heat 6 of them! You are correct that 2 should draw 600-700ma.
DC offset. Should I be seeing this much (9.4mv) offset at the audio outputs when only the tube heater has power? The +/-24 board power is not connected and is a completely different circuit on the board.
With the internal shield (pin9) grounded or lifted, same 9.4mv on audio outputs. Is this voltage just the extra electrons emitted by the heater/cathode hitting the grid and normal?
Taking a stab here - with no voltage on the anode and the cathode being heated, some electrons are going to migrate to the grids - one of which is connected to your output.
Let the heater warm up and the audio out reads -9.4mv. Multimeter is connected with proper polarity. Quickly turned on +/-24v power supply to board and reading quickly went to +2.7v. Switched off board power supply but left heater supply on and it reverses course and is slowly climbing back towards -9.4mv.
I know tubes can pass dc on startup, but with the heater already warm should it pass a large amount at startup? Anyhow, I'm thinking it should never see 2.7v? No smoke, yet.
Yeah - 2.7v of DC offset on the outputs is very problematic for almost any amp this thing would be feeding! That's the point where I would throw in the towel on this and chalk it up as a learning experience.
Gary Richrath
When I wrote that, I woke up one night, half asleep, and sat down in front of the TV. There was a soap opera on it. I was just sitting there, strumming a guitar, thinking, ‘God, these guys’ relationships are worse than mine.‘ I just sat there and sang vocals about the effects of gossip and relationships breaking up, which was what was on the tube and all that was similar to what was going on in my life.
Been running traces and checking them against the schematic and part pinouts. A pic in the seller's add shows a preassembled board with d2 1n4148 reversed according to my board and schematic. Shot jim's audio an email, no reply yet but running board traces and checking polarity of d2 with pinout of q1 shows the cathode of d2 into pin6 and base of q1, which agrees with schematic and my finished board. That still does not explain why their preassembled board is backwards.
Wondered if a silkscreen had been reversed leading to a backwards part, so I ran all the traces, checked them against the schematic and pinouts of q1/q2/q3/d3 and they all look correct. The BOM matches the schematic, which does not completely rule out transposing an item. I am not able to find a mistake on the board.
Only one thing left to check is the pinout of p1 potentiometer which sets dc offset.
Board supplier's reply "D2's position please follow errata indicated direction. In the recent round of fab we have corrected the silkscreen direction". No reason for their opposed mounting in their pre-assembled pic. So, my board is correct with the cathode of d2 1n4148 facing the tube. Check that off the list.
p1; Bourns #3296w-1-10lf, current of 5ma from diodes d1a/d1b into wiper, the cw side of pot into cathode1, ccw side of pot into cathode2; used to adjust dc offset at the audio output. That is as is on board and appears to agree with schematic. Check.
No reply from mosfet supplier, so I sent another email asking for him to test or otherwise verify their validity.
I'm green as a persimmon with this electronics stuff, and maybe missing something obvious to someone more skilled, but I'm just not seeing a mistake on the board, wiring, soldering, or power supply.
Hate to blame the mosfet supplier who answered several questions about their product, but it ain't looking good ...
Can someone here test them to verify they are indeed 2sj79 and 2sk216?
Just a suggestion, but another pair of eyes on things can sometimes be helpful. Since Tom seems to know a bit about this and has been your main help, what about sending it to him to inspect it? ....if he's willing to do so
I'm more than willing. I do have one thing for you to try before that. I'm wondering if the 2.9v you saw when everything was powered up is just an offset issue? Since that's governed by the tube, any mismatch between sections could cause that. Maybe throw a 50K resistor across the outputs and see if you can null out that voltage with the pot.
Had a tech from work look at the board. At first he questioned the polarity of cap c4 which has the cathode to the -24v board power supply, but after further examination stated it should be correct with the cathode toward the -24v power supply and anode to ground.
Still at a loss with this problem.
Long story short, been home a lot lately keeping an eye on things because of a meth head. Lemonade from lemons I guess ... now I have time to work on some unfinished projects.
Pulled all the parts off the borberly boards, checked everything, and replaced all the diodes, voltage regs, transistors, etc. that were bad, probably bad, or fake. Got rid of the piggyback currented diodes (5ma) and replaced with a single 4.5ma.
Questions:
Odd thing, when connected to the audio outputs, why does the multimeter give a reading of 40mv dc when no power is connected to the boards? Double checked this with another multimeter.
The tube heater circuit is a separate circuit from the rest of the board, when the heater is powered up the reading at the audio outputs swings up and down until it sort of steadies itself and begins the rise to 65mv, where it stays. When the heater is powered down, it rises to 200+mv and very, very slowly drops back down, but not to zero. Both boards do this and are within +/-.3mv of each other. What causes this? Normal?
On my last pair of those hard to find $ transistors and yet to find an obvious error. Taking it slow.
I'm going to guess you are seeing a small current flowing just due to the fact that you are heating the cathodes. Those electrons are looking for a higher DC potential. They won't all jump to the grid or the plate when there's no power applied to either, but it's possible some are bouncing around the vacuum as they leave the cathode and are just colliding with the other metal elements in the tube, causing the small voltage you are measuring.
That's what I'm guessing, especially considering both boards are within .3mv with the heaters powered. Was going to fire it up and see what smoked next, but spent the day running wiring and adding a motion light.
Two separate power supplies; one 12v smps with a buck convertor set to 6.3v for the heaters, and a linear +/-24 power supply for the boards. I've noticed a problem with powering down. Powering up the heater, and next the board power supply, no problem, but when the switch is flipped to power down the board supply, it triggers the buck converter to reset. Since the buck converter operates only off the smps output, I'm guessing the little smps is letting a spike through its output when the linear power supply is turned off that upsets the buck convertor. If I ever get this thing up and running, I'm afraid this could send dc through the amps.
What would be a simple and effective way of dealing with a surge caused by powering down the board supply so the buck convertor doesn't reset?
That's very strange. The heaters probably draw too much power for a cheap zener. I guess I would try a simple voltage reg, like the LM317 set for 9-10 volts in front of the buck converter. Seems redundant, but it might catch the spike and they are less than a buck.
Comments
Something doesn't seem right. 2 6DJ8s should not be pulling 1.8A. That's enough current to heat 6 of them! You are correct that 2 should draw 600-700ma.
Smaller 10w $3 smps https://www.ebay.com/itm/AC-110V-220V-TO-DC-5V-12V-24V-2A-10A-15A-20A-40A-60A-Switch-Power-Supply-Adapter/151928719853?ssPageName=STRK:MEBIDX:IT&var=451203659760&_trksid=p2060353.m2749.l2649
and fancy $15 buck converter. https://www.ebay.com/itm/0-5-30V-4A-DC-DC-Buck-Boost-Converter-Adjustable-CC-CV-Step-Up-Down-Power-Supply/333900291427?ssPageName=STRK:MEBIDX:IT&var=542961167725&_trksid=p2060353.m2749.l2649
The buck converter looks to be right on the money when checked with a multimeter and doesn't drift, so far. It's fan is noisy. Looks like it shouldn't need a fan with less than 35w so may disconnect it, or, can I put a resistor in line with the fan power to slow it down? As it sits, I like how it helps draw air into the enclosure and reluctant to completely remove it.
DC offset. Should I be seeing this much (9.4mv) offset at the audio outputs when only the tube heater has power? The +/-24 board power is not connected and is a completely different circuit on the board.
With the internal shield (pin9) grounded or lifted, same 9.4mv on audio outputs. Is this voltage just the extra electrons emitted by the heater/cathode hitting the grid and normal?
Throw me a bone guys ... nothing on seeing 9.4mv on the audio outputs with only the heaters powered?
I wish that I could help, but I honestly don't know.
Taking a stab here - with no voltage on the anode and the cathode being heated, some electrons are going to migrate to the grids - one of which is connected to your output.
Let the heater warm up and the audio out reads -9.4mv. Multimeter is connected with proper polarity. Quickly turned on +/-24v power supply to board and reading quickly went to +2.7v. Switched off board power supply but left heater supply on and it reverses course and is slowly climbing back towards -9.4mv.
I know tubes can pass dc on startup, but with the heater already warm should it pass a large amount at startup? Anyhow, I'm thinking it should never see 2.7v? No smoke, yet.
Yeah - 2.7v of DC offset on the outputs is very problematic for almost any amp this thing would be feeding! That's the point where I would throw in the towel on this and chalk it up as a learning experience.
That's hard to do.
"Take it on the run baby"
Gary Richrath
When I wrote that, I woke up one night, half asleep, and sat down in front of the TV. There was a soap opera on it. I was just sitting there, strumming a guitar, thinking, ‘God, these guys’ relationships are worse than mine.‘ I just sat there and sang vocals about the effects of gossip and relationships breaking up, which was what was on the tube and all that was similar to what was going on in my life.
Gary and Kevin wrote some killer songs. One of my favorite bands growing up.
Relax, all right? My old man is a television repairman, he's got this ultimate set of tools. I can fix it....
sorry for the derail KB 😬
Doesn't that make it OUR time.?
Mr. Hand
"Oh NOOOOOOOOOOO!!" - Mr. Bill
InDIYana Event Website
Been running traces and checking them against the schematic and part pinouts. A pic in the seller's add shows a preassembled board with d2 1n4148 reversed according to my board and schematic. Shot jim's audio an email, no reply yet but running board traces and checking polarity of d2 with pinout of q1 shows the cathode of d2 into pin6 and base of q1, which agrees with schematic and my finished board. That still does not explain why their preassembled board is backwards.
Wondered if a silkscreen had been reversed leading to a backwards part, so I ran all the traces, checked them against the schematic and pinouts of q1/q2/q3/d3 and they all look correct. The BOM matches the schematic, which does not completely rule out transposing an item. I am not able to find a mistake on the board.
Only one thing left to check is the pinout of p1 potentiometer which sets dc offset.
Grrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr
Board supplier's reply "D2's position please follow errata indicated direction. In the recent round of fab we have corrected the silkscreen direction". No reason for their opposed mounting in their pre-assembled pic. So, my board is correct with the cathode of d2 1n4148 facing the tube. Check that off the list.
p1; Bourns #3296w-1-10lf, current of 5ma from diodes d1a/d1b into wiper, the cw side of pot into cathode1, ccw side of pot into cathode2; used to adjust dc offset at the audio output. That is as is on board and appears to agree with schematic. Check.
No reply from mosfet supplier, so I sent another email asking for him to test or otherwise verify their validity.
I'm green as a persimmon with this electronics stuff, and maybe missing something obvious to someone more skilled, but I'm just not seeing a mistake on the board, wiring, soldering, or power supply.
Hate to blame the mosfet supplier who answered several questions about their product, but it ain't looking good ...
Can someone here test them to verify they are indeed 2sj79 and 2sk216?
This is like a frelling soap opera.
Just a suggestion, but another pair of eyes on things can sometimes be helpful. Since Tom seems to know a bit about this and has been your main help, what about sending it to him to inspect it? ....if he's willing to do so
I'm more than willing. I do have one thing for you to try before that. I'm wondering if the 2.9v you saw when everything was powered up is just an offset issue? Since that's governed by the tube, any mismatch between sections could cause that. Maybe throw a 50K resistor across the outputs and see if you can null out that voltage with the pot.
Did you find this thread from 2003? I've only scanned the first post, but he was having issues getting it null.
https://diyaudio.com/forums/solid-state/22883-borbely-low-voltage-hybrid-tube-mosfet-opamp-line-amp.html
Had a tech from work look at the board. At first he questioned the polarity of cap c4 which has the cathode to the -24v board power supply, but after further examination stated it should be correct with the cathode toward the -24v power supply and anode to ground.
Still at a loss with this problem.
Long story short, been home a lot lately keeping an eye on things because of a meth head. Lemonade from lemons I guess ... now I have time to work on some unfinished projects.
Pulled all the parts off the borberly boards, checked everything, and replaced all the diodes, voltage regs, transistors, etc. that were bad, probably bad, or fake. Got rid of the piggyback currented diodes (5ma) and replaced with a single 4.5ma.
Questions:
Odd thing, when connected to the audio outputs, why does the multimeter give a reading of 40mv dc when no power is connected to the boards? Double checked this with another multimeter.
The tube heater circuit is a separate circuit from the rest of the board, when the heater is powered up the reading at the audio outputs swings up and down until it sort of steadies itself and begins the rise to 65mv, where it stays. When the heater is powered down, it rises to 200+mv and very, very slowly drops back down, but not to zero. Both boards do this and are within +/-.3mv of each other. What causes this? Normal?
On my last pair of those hard to find $ transistors and yet to find an obvious error. Taking it slow.
I'm going to guess you are seeing a small current flowing just due to the fact that you are heating the cathodes. Those electrons are looking for a higher DC potential. They won't all jump to the grid or the plate when there's no power applied to either, but it's possible some are bouncing around the vacuum as they leave the cathode and are just colliding with the other metal elements in the tube, causing the small voltage you are measuring.
That's what I'm guessing, especially considering both boards are within .3mv with the heaters powered. Was going to fire it up and see what smoked next, but spent the day running wiring and adding a motion light.
Two separate power supplies; one 12v smps with a buck convertor set to 6.3v for the heaters, and a linear +/-24 power supply for the boards. I've noticed a problem with powering down. Powering up the heater, and next the board power supply, no problem, but when the switch is flipped to power down the board supply, it triggers the buck converter to reset. Since the buck converter operates only off the smps output, I'm guessing the little smps is letting a spike through its output when the linear power supply is turned off that upsets the buck convertor. If I ever get this thing up and running, I'm afraid this could send dc through the amps.
What would be a simple and effective way of dealing with a surge caused by powering down the board supply so the buck convertor doesn't reset?
That's very strange. The heaters probably draw too much power for a cheap zener. I guess I would try a simple voltage reg, like the LM317 set for 9-10 volts in front of the buck converter. Seems redundant, but it might catch the spike and they are less than a buck.