More purple! Ha ha! I prefer mellow amber, and I really dislike retina-burning blue.
Maybe I'll give friction fit a try. I don't have a drill press, which would be helpful. I suppose moving up in drill sizes in conservative steps will work. Adding a dab of 3M Super 77 might also help.
Resourcefulness, creativity, ingenuity: that's all it takes, apparently.
Maybe I'll give friction fit a try. I don't have a drill press, which would be helpful. I suppose moving up in drill sizes in conservative steps will work. Adding a dab of 3M Super 77 might also help.
Resourcefulness, creativity, ingenuity: that's all it takes, apparently.
You may add a little experience here and there, and some guidance from great helpful folks (that is, like mark, Salas, ZM, Nelson, and so many more
Last edited:
for blue, especially clear ones, allowing more than 0.5mA is resulting in , as you said, retina-burning
it's easy - know your DC voltage, minus 2V8 for blue LED, divide that with 500uA and you get resistor value
example for 22V5 rail ( as in F4) (R=U/I, [Ohm,Volt,Amper]
R=(22V5-2V8)/500uA=39K4 use first close standard value, so 39K
dissipation in resistor is P=I^2 * R = 500uA^2 * 39K = 0.00975W , so even tiniest resistor will do
it's easy - know your DC voltage, minus 2V8 for blue LED, divide that with 500uA and you get resistor value
example for 22V5 rail ( as in F4) (R=U/I, [Ohm,Volt,Amper]
R=(22V5-2V8)/500uA=39K4 use first close standard value, so 39K
dissipation in resistor is P=I^2 * R = 500uA^2 * 39K = 0.00975W , so even tiniest resistor will do
Attachments
Hei soldersmoker
I didn‘t meant to „lecture“(?) in any way.
Me too owe him and the others some good stuff.
Thanks, guys. I ended up drilling a hole through the middle of my PSU cap board, mounting it on the front panel. I fed the LED through the hole (friction fit in the board) and now it works fine, though I might lower the resistance a bit since the purple LED is slightly dim. It looks pretty good, I have to say.
But speaking of "dim," check out this damage:
Obvious damage is obvious. The coloration on the board is just a reflection of some trees. The board looks fine.
It looks to me as if the tin alloy on the back of the MOSFET is melted at its top. Maybe the MOSFET wasn't cinched down on the keratherm pad properly and overheated? Or was there a short? I don't think there was a short, as the resistor quietly started smoking after everything had passed the dim bulb test. After I plugged the amp directly to the wall, I started biasing both channels and all was fine for about ten minutes when I smelled the sad stink of magic smoke being let out. So, yeah, I suspect overheating.
One channel works perfectly fine. This channel was damaged when I needed to resolder some wires and had to pull the board out (without unbolting the heatsink--LAZY FOOL). I then put the board back on the heat sink and did a bad job of sliding the keratherm pads underneath them (I'm pretty sure). I have three unmatched IRFP9240s from Mouser. Is it not advisable to replace this [most likely] damaged MOSFET with a new unmatched one? Any other gotchas I should be aware of?
My "mono F4" sounds great by the way. My Linear Tube Audio preamp only swings 4V p-p, according to LTA. The one working channel sounded "almost" loud enough with the volume maxed out. I'm certain I'll need to use my Darling 1626 SET amp to feed the F4's inputs. I already have a pair of 16-ohm 3W resistors to put across the SET's outputs.
But speaking of "dim," check out this damage:
Obvious damage is obvious. The coloration on the board is just a reflection of some trees. The board looks fine.
It looks to me as if the tin alloy on the back of the MOSFET is melted at its top. Maybe the MOSFET wasn't cinched down on the keratherm pad properly and overheated? Or was there a short? I don't think there was a short, as the resistor quietly started smoking after everything had passed the dim bulb test. After I plugged the amp directly to the wall, I started biasing both channels and all was fine for about ten minutes when I smelled the sad stink of magic smoke being let out. So, yeah, I suspect overheating.
One channel works perfectly fine. This channel was damaged when I needed to resolder some wires and had to pull the board out (without unbolting the heatsink--LAZY FOOL). I then put the board back on the heat sink and did a bad job of sliding the keratherm pads underneath them (I'm pretty sure). I have three unmatched IRFP9240s from Mouser. Is it not advisable to replace this [most likely] damaged MOSFET with a new unmatched one? Any other gotchas I should be aware of?
My "mono F4" sounds great by the way.
My Linear Tube Audio preamp only swings 4V p-p, according to LTA. The one working channel sounded "almost" loud enough with the volume maxed out. I'm certain I'll need to use my Darling 1626 SET amp to feed the F4's inputs. I already have a pair of 16-ohm 3W resistors to put across the SET's outputs.
Member
Joined 2009
Paid Member
Omegas with alnico drivers: these.
I like 'em, but they're a tad "polite." They are very quick and revealing. I need woofers or subs that keep up with them. I'm still thinking about how to do that. I have plans from John Inlow for a tapped horn sub, but I'm thinking I need more mid-bass, not bass-bass. I don't listen to a ton of bass-heavy music, but I do like reggae and some orchestral organ music.
Ouch! Sorry you ran into a problem.
I would advise against simply replacing one mosfet with a random one.
I assume the triplet was Vgs matched so that they will share the current
more or less equally. You want a replacement that will preserve that.
Dang. That's what I feared. I'm inclined to just build an Aleph-J now because I'm less enamored with the idea of running two amps when one sweeting-sounding one would do. . . . I guess I can try and measure the Vgs in situ if that's possible, but I'd probably have to do some desoldering and that's not appealing, especially if I end up damaging more MOSFETs.
Just saw the steps required to measure Vgs at Vishay's website: NOPE.
Last edited:
Ouch! Sorry you ran into a problem.
I would advise against simply replacing one mosfet with a random one.
I assume the triplet was Vgs matched so that they will share the current
more or less equally. You want a replacement that will preserve that.
Dang. That's what I feared. I'm inclined to just build an Aleph-J now because I'm less enamored with the idea of running two amps when one sweeting-sounding one would do. . . . I guess I can try an measure the Vgs in situ if that's possible, but I'd probably have to do some desoldering and that's not appealing, especially if I end up damaging more MOSFETs.
Woo-hoo! I went for the gusto and tried one of the IRFP2940s from my Mouser "six pack." Bias on the relevant resistor was only 185 mV compared to the others' 200, but what did I have to lose? And it's making great music! My Omegas no longer have a mid-bass suck out.
Volume is plenty loud in stereo. But I'm keen to try the SET in the chain now. Thanks, everybody!
Volume is plenty loud in stereo. But I'm keen to try the SET in the chain now. Thanks, everybody!
Okay, so I just powered up for the first time to test the power supply. F4 boards are not connected. Used the light bulb test and that part worked fine. However, one LED flickered and then went out. That leads me to believe the LED is oriented properly, but maybe it just has a bad solder joint? Or is there such thing as bad LEDs? I'm getting a steady 25.5v on the rails. I ran into the same issue with my Aleph J, and asked basically the same question at the time on that thread, so I think I understand that as long as the voltage on the rails looks good, whether or not the PSU LEDs light up is a matter of personal aesthetic choice. It's not strictly necessary for the proper functioning of the amp. Is that correct? Just want to confirm before I proceed!
Thanks, ZM! Now to decide whether to futz with it or just keep moving forward... I have a feeling I won't be able to sleep quite right with that LED not working. Which brings me to a question: I clearly have a hard time deciphering the polarity of the LED placement on the PSU from the markings on the PCB (I'm using the Universal Power Supply boards). And I don't see an LED on the PSU schematic for reference. Any recommendations for getting it right the first time?
Yes, it's a 10Kohm resistor. Same as the other side of the PSU, on which the LED is working fine.