Millett "Starving Student" hybrid amp - Page 260

Scott,

I tried it without the heatsink on one of the PCB prototypes and it ran very hot. Maybe functionally ok, but too hot to touch. I used a different type of heat sink to deal with this. My photo hosting is down right now so I can't post a picture but it was an adhesive backed type which can be found at Mouser and is not all that expensive and is certainly a lower profile than the "stock" sink in Tom's (excellent) BOM.

oh great, i'd love to see that.

It's all a question of the mass of the heatsink (and the surface area). These MOSFETs are dissipating a ton of power and need some pretty serious heatsinking. You could probably go with something smaller than the recommended sinks but not too much. A large aluminum case might be ok, although personally I'd stick with heatsinks.

What's the latest on power supplies for this beast? Any great deals going on for P2P wired version? The last I saw was quite a few months ago, and the power supply from the original BOM was still the one to get.
Life has been chaos in the bhjazz house, but I hope to have some time coming up soon!

Well, now that you ask... if you (or anyone else) are feeling somewhat adventurous and would like to test an idea... I just finished (in the drawing board) a Millett "Starving Student" Hybrid version using an alternate tube and power supply. The 12AE6 tube is used to substitute for the 19J6 and the power supply unit is an HP 30 volt@400 mA supply from the Deskjet 600 series ( HP C2175A ). A friend of mine in the US told me these supplies are plentiful and cheap, and the 12AE6 is far easier to find than the 19J6 currently.

Unfortunately I don't yet have neither the tubes nor the supply to actually build this version. I'll order from Beezar and e-bay this weekend but it'll be a good 2 weeks until I get the parts. I haven't been able to find a 12AE6 SPICE model to use in creating a simulation in TINA-TI either. Still, I'm pretty sure (90%) that it'll work. My only worry is how noisy the supply is and/or if it sags badly under load.


The required changes to Pete Millett's original circuit are:

R1/R7 = 22 Kohms
R5/R11 = 2.2 KOhms

If my calculations are right these should set the operating point of the tube at around 19 volts plate voltage (with respect to the cathode) and 0.42 mA of plate current.


R2/R8 = 330 Kohms
R4R10 = 390 Kohms

These ones should set the MOSFET gate at 16 volts.


The rest of the circuit components remain unchanged.

The advantage of using the 12AE6 is that it was designed to run at low plate voltages (rated for 30 volts max) and thus, I think, should be more linear that other tubes which were designed/built for working at 100 volts plate or so. Indeed, as I'm sure most of you know, it's the tube Pete Millett used for the hybrid design he published in Audio Xpress magazine.

The advantage of using a 30 volt supply with the 12AE6 comes from the fact that we're dealing with a 12 volt heater tube. Thus the MOSFET's would be taking around 17.4 volts @ 150mA which is a little less power than they take in the original 19J6 version. That way you don't have to go nuts over the heatsink requirements for the MOSFET's.

Incidentally, a 12A_7 version could be built using this HP supply as well, and the above values for R2/R8 and R4/R10, but I have no way to choose appropriate values for R1/R7 and R5/R11 at such low voltages. It could be that the same values work ok.

I'm sure there are many in this forum far more experienced than I am in the use of this tube and/or supply so, if you have any comments, suggestions, corrections, chime in!


cheers!

Well I finally got around to finishing a build using 12SR7GT tubes as logistic posted a while back, sounds very nice, quite similar to the 19j6, and the tubes were super cheap, 4 bucks each I think they were.

My camera and my picture taking skills both suck so excuse the crappy pic, I redid my original 19j6 build that I put in a cigar box for the 12sr7 since I had built a second 19j6 one in a nice little hammond case.

The heatsinks do get pretty warm, but I'm using 2" tall ones and I'm not concerned that they are running too hot, they pass the 5 second finger test.

Hi, I've come here looking for help with my Starving Student build: I just completed it (I used the PCB, but left out some optional resistors and caps, I jumpered where necessary), and nothing seems wrong with it (no short circuits or anything that I could see), but when I turn it on, the tubes start glowing and they kinda pulsate and when I plug in the earphones (using cheap iPod ones) there's a "clack" noise coming out in the same rhythm as the tubes pulsate... Anyone know what I might have done wrong? Any help is greatly appreciated!

It sounds like you have something wired wrong. Is it possible you jumpered a cap? That would not be good. Capacitors are only meant to pass AC (the music signal). If you deleted some caps, but put jumpers in instead, that would be bad news. Resistors should be jumpered, on the other hand.

If this isn't the problem, we need some pics to see what's going on.

No, I only jumpered resistors (R14-17)... The periodic clicking makes me think something's wrong with a cap too... I'll take pictures next time I'm at my friend's, cause I'm doing everything there since I don't have a soldering iron.

OK, we'll wait for the pics. Meanwhile, the only thing I can think of is something that would affect the entire circuit - have you checked the power connection? Remember the center post of the power socket should be positive (B+ on the PCB). Just a guess ...

Yeah I'll just try switching the polarity on the power socket next time I'm there.

Try measuring it or inspecting it in detail, first. Don't just switch it to try something different. I'm only taking a guess - we don't know that reversing the polarity won't blow something up if you happen to have it correctly now.

if you want to email me the pic, i can host it for you.

Well, I went to my friend's again and switched the wires, and it works now! The only problem left is that on one side, I only hear the music very distantly and it's extremely quiet... I guess something in the other channel isn't soldered properly or something like that...

Good work! Actually, my SS is done, just need to seal up the case and plug it in so I won't be able to make any circuit changes on this one. Please post when you get your updated version working. Its cool to see so many different offshoots from this design.