Millett "Starving Student" hybrid amp - Page 228

I can't really follow this one. The MOSFET is insulated from the heatsink right (or needs to be)? And IF you had conductivity between the heastink and mosfet, as long as it doesn't touch anything else of the components that doesn't really matter.

You are correct since your heatsinks are not tied to ground (of which I was not aware). My point was to make sure the insulator for your mosfet is installed/working properly. I've installed devices and had the insulator not work. In general this is a bad thing. It only takes a few seconds to check and is a good habit in general.

It's not your fault, just me not looking closely at what I actually needed. Ah well, such is life.

So I've just finished my 2nd millet starving student build, this time using a PCB. Want to thank everyone here for their help, especially for the effort of supplying these pcbs

Sounds great and a lot quieter than my messy P2P build...however the balance seems quite a bit off...got to look into that one. Other than that it sounds fantastic and I've graduated from plastic maplin boxes lol.

Very clean look, gurusan. Congratulations!

Do you have any shots showing how you attached the MOSFETs?

Thanks

Inside is not as pretty but not too bad either imo. I used arctic silver 5 epoxy that I had laying around to bond the mosfets to the top of the chassis (insulator in between).

Also the amp has been playing Metallica, Sonata Arctica, and various electronic music for over an hour now and I just checked the mosfets with a temp probe and they are 46-48C so the heatsink and chassis seem to be doing about 15C better than my old to-220 sinks.

Nice build I think I'm going to make both the PCB and p2p, I've got the board and the tubes. Probably just to make sure I got one working amp (the pcb) when I mess up my p2p hehe

Tob bad about the imbalance, the other guys will probably find it though

thanks, I think the imbalance thing might be a grounding problem? Or perhaps the ghetto way I hooked up the pot.

Looks good!

My vote is for the pot wiring. Although, it's difficult to tell from the pic, but is there a chance that R16 and R17 are different? I think I see a different band on one of the resistors compared to the other one.

Perhaps that's the reason...I did not have any 50k resistors so wired 2 33k in series for 66k. I figured that would just slightly reduce the gain though unless I misunderstood the comment for those parts on the BOM?

However, I now have 2 51k resistors, though I haven't wired them in yet.

They were probably for a different channel thus the higher gain one was from the channel that had the least sound. I guess...

With a 50K pot, yes - the 2 x 33K resistors would result in a slight gain reduction. Maybe it's just the wiring, then?

Have you tried swapping tubes? A channel mis-match is possible. Even though I test them all, I have made mistakes and still do. If the channel mis-match follows the tube swap, then that would be the problem.

I'd think a tube would have to be failing to cause a channel mismatch that would be perceptible in this amp.

I don't particularly like the look of the melted Wima on the output although I'm not sure it's damaged enough to matter.

I've swapped tubes and the problem is still there. Going to double check my pot wiring tomorrow and replace that wima cap. Got tired with the soldering iron and melted it...several times lol.

haha. It really amazes me how the minds work for many of the engineers and gurus that frequent this forum.

I see a populated pcb and think to myself "wow, that looks cool" then I move on. The gurus come in and they spot melted caps, backward diodes, inefficient wire gauge, shorted traces, .....