I can't seem to get this board going.
I am using 45 for output tubes and when I plug the rectifier tube in my Mosfet gate and source voltage go to over 100v positive. Without the rectifier I get the correct -150 rail voltage and my bias voltage to the 45 are -60v and adjustments can be made via the pot. But as soon as I plug in the rectifier and B+ ramps up I get Vg=112 and Vs=114 on the Mosfet. I only leave it like this for a few seconds because the 45's are pulling over 100mA each. Adjusting bias pot does nothing. I have a 190-0-190 power transformer that when the tubes are pumping current the 45's Va=180.
Are the Mosfet's gonzo? They are NDF02N60Z. I was told to use these type. The previous owner of the board had the same issue with the same Mosfets. I have some FQPF2N70 that I want to try if I can just get the blessing in that it's okay to do so.
So I believe I narrowed it down to the Mosfets, is there anything else to look at? I checked grounds and traces and everything looks fine. I got the board used because the original owner couldn't get it working either. When I got the board I stripped all the parts off and started over, I took my time and replaced a good majority of parts except for resistors that checked good and the 6.3v caps. All other caps are new.
I am using 45 for output tubes and when I plug the rectifier tube in my Mosfet gate and source voltage go to over 100v positive. Without the rectifier I get the correct -150 rail voltage and my bias voltage to the 45 are -60v and adjustments can be made via the pot. But as soon as I plug in the rectifier and B+ ramps up I get Vg=112 and Vs=114 on the Mosfet. I only leave it like this for a few seconds because the 45's are pulling over 100mA each. Adjusting bias pot does nothing. I have a 190-0-190 power transformer that when the tubes are pumping current the 45's Va=180.
Are the Mosfet's gonzo? They are NDF02N60Z. I was told to use these type. The previous owner of the board had the same issue with the same Mosfets. I have some FQPF2N70 that I want to try if I can just get the blessing in that it's okay to do so.
So I believe I narrowed it down to the Mosfets, is there anything else to look at? I checked grounds and traces and everything looks fine. I got the board used because the original owner couldn't get it working either. When I got the board I stripped all the parts off and started over, I took my time and replaced a good majority of parts except for resistors that checked good and the 6.3v caps. All other caps are new.
In the Tubelab SE: Removing MOSFET's thread, a couple down, Tomchr, in post #5, describes how to bypass the MOSFET's.
I haven't tried it, but you might give that a try to see if that brings your board to life.
Also, the center pin of one of the 10M45's is used to pass voltage between the top and bottom sides of the board; if the solder doesn't plate through the hole, shenanigans occur. It's described on the last page of the 801A in Tubelab SE thread a page or two back.
Win W5JAG
I haven't tried it, but you might give that a try to see if that brings your board to life.
Also, the center pin of one of the 10M45's is used to pass voltage between the top and bottom sides of the board; if the solder doesn't plate through the hole, shenanigans occur. It's described on the last page of the 801A in Tubelab SE thread a page or two back.
Win W5JAG
I believe this is the thread you are talking about.
http://www.diyaudio.com/forums/tubelab/291593-tubelab-se-removing-mosfets.html
http://www.diyaudio.com/forums/tubelab/291593-tubelab-se-removing-mosfets.html
You are a saint! I bet this is what is happening, the traces are gone on the bottom of board, I thought since the top was making a connection it was okay but it appears to be a vias.
I will fix this and report back!
I see a trace on top side of board connecting the drain of the 10M45's on both channels to B+, there is also a trace on the bottom of the board connecting the drains of the MOSFETs for both channels connecting to B+. Beep test confirms visual inspection of connected traces.
The bottom pads for the 10M45 drains are lifted off but I don't see a trace on the bottom where it would connect to anything. Like I said everything that is supposed to be connected to B+ beep tests continuity fine.
An externally hosted image should be here but it was not working when we last tested it.
Starting to think nobody looks at this section of the forums, including George.
I agree and I did that with my heater power question but the powers that be moved it here where nobody looks, awesome.
Well I checked everything a million times, negative rail is perfect without rectifier , I plug in the 5AR4 (no other tubes in) and the Mosfets Gate and Source ramp up to over 100v positive. Drains are all good connections to B+, Sources have 3 watt metal oxide 22k to negative rail and 100R grid stop for 45, gates are connected to to their respective coupling caps and bias networks.
It's not rocket science here so I must move on and start replacing things. I am going to start with the Mosfets and I will report back.
It's not rocket science here so I must move on and start replacing things. I am going to start with the Mosfets and I will report back.
Good grief.
It's been less than twenty four hours since your first post. Not everyone who looks at this forum is an expert long distance troubleshooter.
Was bypassing the mosfets as was suggested in the earlier thread unsuccessful?
Win W5JAG
Hahaha I know I am a PITA!!!!!!!!!!!! BUT you must admit the tubes/valves section would most likely have had more views and likely more posts. Yeah I know my life is pretty sad if it revolves around building tube amps and diyaudio
So here is an update:
I finally just said it must be the Mosfets especially since I found the new ones in a box that I thought I installed. I removed the ones that were installed and turned it on with a 5AR4 and everything looked stable now. I installed the Fairchild parts and all seems fine now, it was the Mosfets after all.
I had a late night of soldering and I must have forgot to fit new Mosfets on the heatsink before calling it a night, I left all the semiconductors in a pile for the next day to get installed on the board so the subsequent day I went down and just plugged em in too excited to hear this thing. It cost me an extra day of debugging for one minor mistake. DUH!
Thanks for your help W5JAG and wdecho.
Lose bird
So here is an update:
I finally just said it must be the Mosfets especially since I found the new ones in a box that I thought I installed. I removed the ones that were installed and turned it on with a 5AR4 and everything looked stable now. I installed the Fairchild parts and all seems fine now, it was the Mosfets after all.
I had a late night of soldering and I must have forgot to fit new Mosfets on the heatsink before calling it a night, I left all the semiconductors in a pile for the next day to get installed on the board so the subsequent day I went down and just plugged em in too excited to hear this thing. It cost me an extra day of debugging for one minor mistake. DUH!
Thanks for your help W5JAG and wdecho.
Lose bird
I am also having trouble getting my Tubelab SE board going. I am using 45s, an Edcor XPWR221, 500VCT transformer, and all parts are as specified in the parts list. I am getting only 15-18 volts at the positive lead of the coupling cap (varying with the trim pot settings) closest to the 5842s. Same on each channel.
B+ is 340 with the GZ34 in circuit.
Any ideas why I am seeing 1/10th the expected voltage at the coupling caps?
B+ is 340 with the GZ34 in circuit.
Any ideas why I am seeing 1/10th the expected voltage at the coupling caps?
Last edited:
Problems with Tubelab SE checkout
"Next, put in the 5842’s. Put a meter from the plate of each 5842 to ground. The plate is easiest to access by clipping the positive meter lead on to the coupling cap lead closest to the 5842. The cathode pots should adjust the plate voltage on the 5842’s. Set them to about 175 volts (not critical). The pots were added because of the wide variation in 5842’s that I saw. Power OFF the board. Do NOT touch the board for 5 minutes after the power is disconnected."
That is what I was measuring (plate voltage on 5842s), but I'm seeing 1/10th the voltage that I should on each tube.
"Next, put in the 5842’s. Put a meter from the plate of each 5842 to ground. The plate is easiest to access by clipping the positive meter lead on to the coupling cap lead closest to the 5842. The cathode pots should adjust the plate voltage on the 5842’s. Set them to about 175 volts (not critical). The pots were added because of the wide variation in 5842’s that I saw. Power OFF the board. Do NOT touch the board for 5 minutes after the power is disconnected."
That is what I was measuring (plate voltage on 5842s), but I'm seeing 1/10th the voltage that I should on each tube.
Does adjusting the bias pot (at the cathode of 5842) change anything?
The adjustment pot is parallel with the cathode resistor and there is no other fixed resistor in series with it to stop it from creating a very low cathode resistance. I.E. if the adjustment pot is set for too low a resistance, say 10 ohms, 10*.01=100mV or .1v Since the grid is referenced to ground you will have Vg-k= -.1 and so the tube is biased too far to the left of the curves and can give a very low plate voltage. Try to increase the cathode resistance to ~200R, this should bring your bias voltage up to 200*.01=-2v which will give you closer to 150v at the anode.
The adjustment pot is parallel with the cathode resistor and there is no other fixed resistor in series with it to stop it from creating a very low cathode resistance. I.E. if the adjustment pot is set for too low a resistance, say 10 ohms, 10*.01=100mV or .1v Since the grid is referenced to ground you will have Vg-k= -.1 and so the tube is biased too far to the left of the curves and can give a very low plate voltage. Try to increase the cathode resistance to ~200R, this should bring your bias voltage up to 200*.01=-2v which will give you closer to 150v at the anode.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.