Millett "Starving Student" hybrid amp
Dec 26, 2008 at 7:06 AM Post #1,801 of 7,277
Quote:

Originally Posted by krisjan /img/forum/go_quote.gif
but then, why are input caps (on other amps) so low on capacitance


Quote:

Originally Posted by Logistic /img/forum/go_quote.gif
My understanding of electronics is pretty limited though, so someone else might be able to give a better explanation.


Basically correct. For some more detail, see ecp.cc
 
Dec 26, 2008 at 5:27 PM Post #1,802 of 7,277
Why has nobody else gone out and modified the circuit to use other tubes? The original tube is pretty spiffy with its 18V heaters, but there is no reason that various parts could not be adjusted to use the same ideas with a 12au7 or similar. $25*2 for tubes to put into an amp with $60 in parts?

I have an alternate schematic, using a tube you may ALREADY have from a previous build, for anyone brave enough to try it.
Quote:

Originally Posted by Punnisher /img/forum/go_quote.gif
Anyone know the secret to making this amp more Grado friendly?


link1
Works well with REALLY cheap attenuators, especially if you need LOTS of attenuation.
link 2
Works well with better attenuators, have not tried with cheap parts: I was too happy with the first one.
 
Dec 26, 2008 at 6:16 PM Post #1,803 of 7,277
I had done that to use 12au7 (5963) since I've got a few dozen extras. But, then I figured why bother, and B+ shot up to 300V for a plate voltage of around 220V @ 4-8mA with CCS in a few places....and as all things DIY everything else started changing too, with a PS instead of wallwart.

If you're going to go 9-pin, I would definitely start using some CCS for more tube flexibility. You know that nobody is going to stop with a 12AU7.
 
Dec 26, 2008 at 7:28 PM Post #1,804 of 7,277
well i have followed the schematics exactly built it ( with the odd mishapp in drilling my case) and nothing. spent past 2 days resoldering and de soldering everything none of my four tubes are lighting up or anything. ive prolly killed small bit or something
 
Dec 26, 2008 at 7:49 PM Post #1,805 of 7,277
Quote:

Originally Posted by holland /img/forum/go_quote.gif
If you're going to go 9-pin, I would definitely start using some CCS for more tube flexibility. You know that nobody is going to stop with a 12AU7.


I agree, a plate-CCS will be mandatory, but its really not that hard to build a simple one. Restricting the design to a B+ of 48V pretty much guarantees that you can use low power transistors with high Hfe.
There have been a few commercial products through the years that have used 6dj8 and 12au7 (amongst others, but who *needs* more than 30 voltage gain...) with as little as 30V on the plate. I agree that according to the plate curves it should suck hard enough to get a golf-ball through a garden hose... but isnt that the fun of designing with restrictions? The challenge of who can make the "best" compromises to get the most sound out of junk-bin parts on a lunch-money budget.
Quote:

Originally Posted by Yaka /img/forum/go_quote.gif
well i have followed the schematics exactly built it ( with the odd mishapp in drilling my case) and nothing. spent past 2 days resoldering and de soldering everything none of my four tubes are lighting up or anything. ive prolly killed small bit or something


Is this your first build?
Tube sockets are numbered as viewed from the bottom.
got pics?
 
Dec 26, 2008 at 10:25 PM Post #1,807 of 7,277
Quote:

Originally Posted by OldKing /img/forum/go_quote.gif
I built this one for a friend as a Christmas present.

(photos removed to protect the 56K)



That looks SWEET!
Where did you get the heat-sinks? I really like the look.
 
Dec 26, 2008 at 11:42 PM Post #1,809 of 7,277
Quote:

Originally Posted by OldKing /img/forum/go_quote.gif
Heatsink is here

BOM (more or less) is here except I changed caps and added the Neutrik jacks.

The knob and heat sink are black anodized aluminum which I sanded and polished.



Ahhh that's how you got the sink to look like that. Very nice
beerchug.gif


Did you use a dremel to polish, or??

I like the idea of giving a MHSS as a gift! I'm too late for xmas this year, but if I start now, I might have some ready for next year
tongue.gif
 
Dec 26, 2008 at 11:46 PM Post #1,810 of 7,277
Quote:

Originally Posted by nikongod /img/forum/go_quote.gif
but isnt that the fun of designing with restrictions? The challenge of who can make the "best" compromises to get the most sound out of junk-bin parts on a lunch-money budget.


That's very true. I'm not sure how much can be treaded though, so much has already been done in the <= 48V range. SOHA, SOHA 2, Millett Max, Borbely's hybrid, YAHA, YHHA, MAHA, etc.

To be honest though, every VAS I come up with, for this simple amp, seems to look like the SOHA, SOHA 2, or Borbely. There's been alot of ground tread already and it's going to be difficult to really differentiate.

Heatsink:
641K
 
Dec 27, 2008 at 2:27 AM Post #1,811 of 7,277
Quote:

Originally Posted by nikongod /img/forum/go_quote.gif
Why has nobody else gone out and modified the circuit to use other tubes? The original tube is pretty spiffy with its 18V heaters, but there is no reason that various parts could not be adjusted to use the same ideas with a 12au7 or similar.


12au7, 12sn7, 12j5 all might work, though at the low voltages clipping from overdriving the grid is a real issue.

Here's the version I drew up a while ago. It is higher voltage, but is the same basic topology as the original with the addition of a plate CCS and a shunt regulator. I've been sitting on it as I have not had a chance to build it to see if it works at all. But, I don't think I will, so this should be considered untested.

407a_headamp_original.jpg
 
Dec 27, 2008 at 3:54 AM Post #1,813 of 7,277
interesting, dsavitsk.

Here's one I put together. I intend to build it, the voltages were chosen based on transformers I have.

PS will be a 120V->48V (24-0-24) -> 240V using 2 transformers. 240V rectified to about 330V will be filtered with CRCLC. (24-0-24) will be simple 3 pin regs. I intend to put an e12 on the output, hence no cap and trimpot to adjust. e12, in case of thermal drift, will make it super sensitive to DC. startup delay of about 1 minute for PS to settle.

It's not starving student, and not quite beginner voltages.

I was sitting on this as well, will likely try a build in a few months. I owe some amps to some relatives, so why not, as I already have most of the parts on hand.
 

Users who are viewing this thread

Back
Top