1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.

    Dismiss Notice

Millett "Starving Student" hybrid amp

339 340 341 342 343 344 345 346 347 348
350 351 352 353 354 355 356 357 358 359
  1. Rangiroa
    A little update to my problems from a few days ago.

    I started prodding around with my DMM and found that pin 4 on one tube socket is receiving 13.9 volts, and one the other one its receiving the full 48.

    What connection is missing that would involve allowing all 48v into that pin?
    Also, What is the best way to ground all the ground wires? I attempted to use the Terminals that were for the PCB. I locked down every ground cable into the terminal then ran a wire and soldered it to all of them and screwed that wire to the case.

    Where can i find those things a lot of people are using that look like its a bracket for soldering ground leads too.
    is it proper to ground to my case? I think that might be why my screws spark and snap lol
    Sorry for these basic questions, I appreciate any help on basic electricity principles :)
  2. the_equalizer


    Grounding the metal case is a good practice, as it provides shielding from electromagnetic interference to the the circuit within. Your description of your ground wiring sounds quite alright.
    As to your problem of getting 48 v in one of the heaters, I can only imagine that you miswired the tube heater to the MOSFETs pin 2, which is in turn connected to the 48 v line, instead of to the MOSFET's pin 3.
  3. the_equalizer

    Zero or very little voltage readings in the circuit probably means there's some big short circuit that causes the power supply protection to kick in and shut down the power
    You need to sit down and carefully and patiently trace down all your connections, step by step, against the schematic.
    Remove the tubes, apply power to the circuit and measure the voltage (with respect to ground) at various points in the circuit: positive terminal of the power connector, the junctions of R2,R4 and R8,R10. Reinsert the tubes and measure those voltages again, as well as the voltage of pins 1 and 4 of the tube.
    Frying a tube is really hard, particularly with the voltages used in the SSMH. One tip: Using different colored wire for each kind of connection might seem like extra work while building but it pays off with great interest if the need to debug arises.
  4. Rangiroa
    Thanks for all your help equalizer. 
    I would post pics but i tried taking some and my old camera cant really show the detail too well.

    Its probably just easier to not :)

    I will check out the mosfet wiring to make sure they are wired correctly.

    the drain on the mosfet was very strange to wire bc the schematic shows both mosfet pin 2's connected, is this correct? What i did was run a wire between both pin 2's and then run 1 wire from there. Should i do this a different way, such as connect both pins to the wima caps and then connect the caps together? 
    this may be why im getting one tube to light up and the other to not, because the mosfets being connected incorrectly.

    However, it still doesnt explain why both tubes would light up to normal brightness when i first started up the amp. I must have dislodged a connections. 
    Will update tomorrow with hopefully more progress.
  5. the_equalizer

    No problem, sorry for taking my time to reply but I was away on holidays for 2 weeks.
    No, connecting the MOSFET's pins '1' and '2' (actually pins 2 and 3 in the diagram) to one another is not correct. The diagram connection you mention between drain and source is just the symbol for the MOSFET and somewhat represents it's internal construction/wiring.
  6. Rangiroa
    so basically, I do not want to run a connection between mosfet pin 2's on each one i just want to connect them seperately?
    This is because each tube is its own channel? Am i correct that the only thing connecting both channels is the Capacitor that is grounded and connects to the 33k resistor?
    edit: the volts on the connection between the 390k resistor and the 33k resistor that connects to the grounding 680uf capacitor and the switch are only 13v whereas the schematic says they should be more like 45ish.
    My connections on the mosfets are as follows: each 2k resistor is hooked to mosfet gate pins.

    I ran a wire from both mosfet drain pins to the 2k resistor connected to the switch. should i physically connect the 2 wires at the switch instead of at the resistor? maybe a missing connection there is causing problems.
    source pin 3 is connected by a wire from tube socket pin 4 as well as a secondary wire off tube socket pin 4 which is running to the + side of the wima caps.
    thats how my mosfets are setup.

    I cannot find fault in other places in my wiring aside from it not being as neat as i would have hoped. 
    are there any other trouble spots i should check the volts on to help narrow the problem down?
  7. the_equalizer

    Oh, got it now. Both MOSFET drains (pin 2) are connected to the 48 v line. Theoretically It's OK if you run a wire between the MOSFETs and then run another to the 48 v line, but I'd run an independent wire for each drain to the power line.
    When you say that another wire runs from the tube's pin 4 to the WIMA cap (C3a, I assume)... the electrolytic cap (C3) is also connected there, right ?
    13 volts in the junction of R1, R2, R13 and C6... that's certainly not right, and suspiciously close to the tube heater voltage; of course it might be just a coincidence but I'd re-check that I have not wired the MOSFETs drain and source in reverse.
  8. Rangiroa
    Sorry in advance for the poor quality, but i have an older 6mp camera and I am not quite sure how to take detailed stills with it. I have always had trouble. If anyone can instruct me in basic camera use, id be appreciated. Maybe then i could take better detailed pics. EDIT: in the middle of posting this I have figured out how to use my camera :S

    Anyways, I am desperate to get this thing working. I have a couple pics to attach and hopefully they will be able to describe my problem.
    The first pic here is a pic of my tube wiring on the socket which does not light up. The blue and orange cable in the pic refers to wires that are being grounded, or bridging pins 3 and 8. green wires are connection wires and have been used to connect the components.
    This next picture show a look at how my power switch is connected. Since quality is bad, I would like to note that pin 1 of the swtich has the blue and orange power, and off of pin 2 are 2 wires that are connected to pin 1 of the mosfets (I switched pin 1 gate and 2 drain of the mosfets since Equalizer suggested the voltages seemed as though they were switched) as well as the 2k resistor and the 680uf capacitor. My power switch sparks sometimes when I turn it on.
    These next few are for reference. As i was taking this pics I seemed to have damaged one of the resistors in this next pic so i will be ordering a replacement few. That was fine before just now and not a reason for unworking tube socket i think. CIMG3674.jpg
    This next picture is my working tube socket wiring
    Here are the mosfet connections for the tube that lights up.
    So besides that one resistor that literally broke as i was taking pictures, do you need to see a photo of anything else?
    At this point the only thing I can think of that I may have majorly overlooked is that the wima cap and electrolytic cap are wired together and connected backwords for the nonworking tube, but taht doesnt explain the funny voltages IMO.
    Other ideas of mine are that the power switch needs replaced, or I possibly in one of the resistor connections made a connection at the wrong spot on the connector. I didnt think it would matter as long as they were wired in series like the schematic calls for. Thoughts, Ideas? I will place an order tomorrow for some new resistors. I also accidently ordered 1 watt 390k resistors instead of 1/4 watt. But that still doesnt explain one working tube and one nonworking tube. I would just like to get both tubes fired up and then I will be able to handle the rest i think .... lol.
    Im racking my brain and I cannot find a connection problem, so maybe someone else can here.

    Thanks in advance.
  9. ArtemF
    it's not clear from the pictures if the the MOSFETs are electrically insulated from the radiator/ground.
    If they are not then you ground the sink that for this MOSFET means short-cutting the power supply.
    May be all the troubles started with that, the MOSFETs were fried and everything got messed up?
  10. the_equalizer
    Indeed your MOSFETs are miswired; the gate is NOT the middle pin, it is the first one from the left. See here for the correct pinout.
    ArtemF's observation of MOSFET to case insulation is also very important; it's very likely your MOSFETs are fried :frowning2:
  11. Rangiroa

    I will just order a new set of mosfets. The way they are hooked in the picture was only because equalizer suggested maybe the gate and drain were hooked backwards, I changed it just for effect.

    I believe I mentioned that before I posted the picture.
    Will the heatsink pads sold on beezar do the trick? i was going to order a pair of them anyways for another build.
    Thanks for helping me narrow the situation down. Note to self:dont hook the mosfets up to the case.

    That explains why they worked for a min on the first try, because they werent fried yet :) :)
    I will order my missing parts and respond back in a few days with results :)
    Thanks again
  12. Rangiroa
    What is humorous to me and maybe others can laugh with me, but I thought "hey, the idea of this amp is to save money so im going to use the case as a heatsink and screw the mosfets into the case".
    Oops. Would have saved me lots of headaches had I known how important insulation was for those mosfets.
    Oh well, I have learned alot with this build and provided i hook up the mosfets to my new heatsinks, with the pads and hardware and it works, I will be very pleased with it :)
    Thanks for all the help in regards to this
  13. ArtemF
    next time when you mount your new MOSFETs, try to "reverse-engineer" your circuit.
    I mean, draw a circuit diagram looking just at your connections. If you get the original scheme,
    congratulations. Then you can take the ohmmeter and check for the shortcuts.
    If you don't find any, power the circuit and check the voltages.
    You might also start with just one channel to avoid the mess with the wires. 
  14. Rangiroa
    Thanks I will definately give that a try when I rewire.
    I feel like i quadruple checked everything over the weekend, because i was looking for that 1 possible connection that was screwing everything up.
    But i will definately try your advice before i power up so i dont fry my mosfets.

    I ordered extras of the mosfets, and heatsink pads too this time lol, that way i dont have to wait on mouser if they blow up.
  15. Darkraver
    Just finished my first DIY amplifier, wanted to show it off here :)
    001.jpg 002.jpg
    003.jpg 005.jpg
    I never realized how small 4" x 4" really was, until I started to solder everything together. Also, I really should think about buying another colour for wire, staring into a blue mess doesn't help in debugging, though luckily everything worked the first time I turned it on.
339 340 341 342 343 344 345 346 347 348
350 351 352 353 354 355 356 357 358 359

Share This Page