Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Millett "Starving Student" hybrid amp
New Posts  All Forums:Forum Nav:

Millett "Starving Student" hybrid amp - Page 438

post #6556 of 6797
Quote:
Originally Posted by alphaman View Post

This is a LONG thread, and many may not have the time to trudge thru it all. Not a bad time, methinks, to post an UPDATED schematic and BOM.

The first post of a forum thread is always the easiest to find. wink.gif  Then there was the PCB version, here: SSMH PCB

 

The 12AU7 variant has been stored for posterity here:

SSMH Variants

 

As for the  BOM, etc., you can find it on that site (except the PCBs are no longer available).  In addition, the very first post in this thread has the original schematic and the original BOM.  Almost nothing has changed from that, except the addition of some bypass caps, differences in volume pots, and your choice of heat sinks.  If you go with different tubes, then as stated, the 12AU7 version is archived above. 

post #6557 of 6797
Quote:
Originally Posted by jarscar View Post

The case being metal isn't a big deal, everything that connects to it will be grounded and in the end your case will be grounded also, the only thing you really need to worry about is shorting the circuit by accidentally touching a solder joint or stray wire to the case. The only thing that gave me any problems was the MOSFETs, I had to brush liquid black tape on the connectors because the leads were a lot longer than the insulators I used. But other than those everything should mount directly to your case. 

 

 

Quote:
Originally Posted by KimLaroux View Post

I believe your grounding scheme is not efficient enough. It's kind of a common problem here, as it's usually a first project for most of us. You have to think of the ground as a return path for different signals, and not as a "neutral" point where everything gets nullified. Ground is nothing more than the opposite of the power rail, after all. A perfectly silent tube amplifier uses a balanced wiring for V+ and Gnd throughout the amplifier, with proper decoupling at each stage.

 

The voltage you get at any single point is in reference to the ground at the other side of this point. If that ground point is noisy, the end result is the same as if your V+ was noisy.

 

Power supply grounds are noisy. Signal grounds are sensitive to noise. Therefore, make sure the power return paths are not going across your signal return paths. 

 

This is why the first recommendation most people have is to tell you to use a Star Ground. Connect all your return paths to a single point. Connect this point to the power input jack using a single wire. Connect the star ground to the chassis at a single point, making sure no current flows trough the enclosure. This is not a car, it's an audio device. The enclosure should be used as nothing more than a shield.

 

I had to rebuild my whole amplifier many times to learn all that. Now it's dead silent.

beerchug.gif

 

Thanks a lot guys, I have already finished the amp on the new metal case. I used the star ground method and surprisingly everything works perfectly from the start and the amp is completly silent. Now I just need to wait for the new usb dac to arrive since I don't have the X-Fi modded anymore.


Edited by Amatsu - 6/14/13 at 10:06pm
post #6558 of 6797

tomb: Thx for summarizing! [FYI: My main ref to this project has always been: http://www.pmillett.com/starving.htm ] So, based on your  succinct summary ...this thread seems ridiculously long...i.e.,  if your two external links (Millet's DIY site pages) are the ONLY major topological changes/ mods to the orig. design... then I'm guessing much of the discussions in THIS ~440-page  thread have been ...  ... about what ... MSSH cosmetics ;) ??

 

IAC ... I'm perhaps most curious about the addition of CCS. Not sure whether KimLaroux's recent mention of it is new to this thread and, in general,  the MSSH design. How does CCS improve sound over orig.? 


Edited by alphaman - 6/15/13 at 7:06am
post #6559 of 6797

In general a CCS will force a tube to have a much flatter load line and make it operate in a considerably more linear manner. Kim is using cathode CCSs which I am personally not as keen on as it means that your don't gain any AC impedance to the plate. An anode CCS provides a very high load to the plate and increases PSRR (power supply ripple rejection) by a factor of 10 even for the most basic CCS. (But you do end up needing high voltage transistors etc. as opposed to standard toolbox stuff in the cathode CCS). Swings and roundabouts...

post #6560 of 6797
post #6561 of 6797

That amp is pretty different to a MHSS and the CCS is controlled by CCdiodes which many people aren't particularly keen on for actual music use. If you're using the original power supply scheme then the minimax CCS scheme http://www.diyforums.org/MiniMAX/MiniMAXccs.php  is a good place to start due to its decent performance and ease of implementation (2 transistors, 2 resistors). The Equalizer was the first (AFAIR) to implement this mod, then I also added it back when my SS operated at standard voltages. Its main effect (that I noticed), was giving me a blacker background (the SMPS is far from perfect) and increased the bass punch a bit. I've never come across a CCS that has completely changed a circuit, when it comes to tubes they generally just seem to be a final layer of gilding (one may therefore deduce that if plated in Rhodium they'll sound much better! :P )

 

If you use a CCS like the minimax's one then you'll always be able to retrofit it like I did, so don't be afraid of just building the amp first using a standard resistor bias.

 

Cheers!

post #6562 of 6797
Quote:
Originally Posted by alphaman View Post

tomb: Thx for summarizing! [FYI: My main ref to this project has always been: http://www.pmillett.com/starving.htm ] So, based on your  succinct summary ...this thread seems ridiculously long...i.e.,  if your two external links (Millet's DIY site pages) are the ONLY major topological changes/ mods to the orig. design... then I'm guessing much of the discussions in THIS ~440-page  thread have been ...  ... about what ... MSSH cosmetics ;) ??

 

IAC ... I'm perhaps most curious about the addition of CCS. Not sure whether KimLaroux's recent mention of it is new to this thread and, in general,  the MSSH design. How does CCS improve sound over orig.? 

I apologize for not knowing where to find it specifically, but there is another variant that used metal tubes.  Maybe someone else will point it out. wink.gif

post #6563 of 6797
If I were feeling adventurous and wanted an interesting tube to use for this design, I think I'd be inclined to look at the 3A5. There will be a few design challenges and changes, but it looks like a good candidate to me.
post #6564 of 6797

Just looked that tube up, wish I'd known about it before building mine with 12AU7s, with some modification those could really sing on the 48V supply

post #6565 of 6797

I haven't seen a schematic for the Bottlehead Crack .And while that is a different style of tube amp than MSSH, I did note that Bottlehead offers a CCS upgrade called "Speedball" for the 12AU7 section:

http://www.bottlehead.com/store.php/products/speedball-upgrade-for-crack-amplifier

 

Will it work with the 12AU7 SSMH variant?

 

EDIT: Here's the C4S Camille Cascode Constant Current Source kit:

http://www.bottlehead.com/et/adobespc/C4S/c4s_kit.htm

... it's an LED-based CCS design from VALVE magazine, Oct 1997 and 1999 here.

...some related info here: http://www.pimmlabs.com/

==================================

 

Quote:
What yields an even more intense experience than Crack? The most hardcore users say it's a Speedball. Speedball is our Camille Cascode Constant Current Source upgrade for the Crack headphone amplifier. The 22k1 ohm plate loads of the 12AU7 input triodes are replaced by C4S loads, as are the 3K ohm cathode load resistors of the 6080 triodes. The PC board for the loads on the 6080 tube has very ample heat sinks to assure high reliability.
 
We consider this a skill level 2 kit, meaning that it is intended for someone who has construction of a kit under their belt (like the basic Crack kit). The new skills you might acquire will be those of stuffing and soldering a PC board, and removing a few components to allow installation of the upgrade. Assembly and installation is an easy one evening session, requiring only the removal of four resistors from the previously assembled Crack kit.

 

 

 


Edited by alphaman - 6/17/13 at 2:39am
post #6566 of 6797
Quote:
Originally Posted by tomb View Post

I apologize for not knowing where to find it specifically, but there is another variant that used metal tubes.  Maybe someone else will point it out. wink.gif

 

You're probably refering to this post by user Logistic. One of these days I have to build one...

 

 

EDIT:  Also, for the more adventurous, you can try building a 12AE6 version

 

 

cheers!


Edited by the_equalizer - 6/19/13 at 7:49am
post #6567 of 6797
Quote:
Originally Posted by Goobley View Post

That amp is pretty different to a MHSS and the CCS is controlled by CCdiodes which many people aren't particularly keen on for actual music use. If you're using the original power supply scheme then the minimax CCS scheme http://www.diyforums.org/MiniMAX/MiniMAXccs.php  is a good place to start due to its decent performance and ease of implementation (2 transistors, 2 resistors). The Equalizer was the first (AFAIR) to implement this mod, then I also added it back when my SS operated at standard voltages. Its main effect (that I noticed), was giving me a blacker background (the SMPS is far from perfect) and increased the bass punch a bit. I've never come across a CCS that has completely changed a circuit, when it comes to tubes they generally just seem to be a final layer of gilding (one may therefore deduce that if plated in Rhodium they'll sound much better! :P )

 

If you use a CCS like the minimax's one then you'll always be able to retrofit it like I did, so don't be afraid of just building the amp first using a standard resistor bias.

Sorry for just getting around to replying to your reply ;)

The amps ain't THAT different in that both are "vacuum tube and discrete solid state hybrid headphone amplifier[s]"

About that diode ... my bad on stating that being the ONLY part needed for a drop-in type CCS upgrade for the MSSH -- and YOUR bad for not noticing that :)

Specifically ... if you look at the  the Borbely ckt you'll note two addit. transistors and three Rs  (Q4, Q5,R11/12/13 )., sorta similar to MiniMax CCS topology, but the borbely uses one MOSFET and one BJT 


Edited by alphaman - 6/19/13 at 10:40pm
post #6568 of 6797
Quote:
Originally Posted by alphaman View Post

Sorry for just getting around to replying to your reply ;)

The amps ain't THAT different in that both are "vacuum tube and discrete solid state hybrid headphone amplifier[s]"

About that diode ... my bad on stating that being the ONLY part needed for a drop-in type CCS upgrade for the MSSH -- and YOUR bad for not noticing that :)

Specifically ... if you look at the  the Borbely ckt you'll note two addit. transistors and three Rs  (Q4, Q5,R11/12/13 )., sorta similar to MiniMax CCS topology, but the borbely uses one MOSFET and one BJT 

 

While it is true this amplifier uses vacuum tubes and solid state devices, the WAY they are used is certainly different. Otherwise one might conclude that all hybrid amplifiers are very similar simply because they use the same basic components... 

 

For starters, the input stage above looks like differential pair with D1A and D1B acting as a tail CCS; a topology similar to the SOHA II. On the other hand the SSMH is a simple common cathode triode voltage amplifier.  

 

The circuit around Q4 and Q5 is a CCS acting as source load for the output MOSFET stage, not for the tube stage. On the SSMH it is the tube heater that acts as the MOSFET load.

 

Finally, you can see that in the above diagram both the tube and MOSFET as well as the output load and MOSFET are directly coupled, thanks to the use of a bipolar power supply, whereas the SSMH uses a simple +48V supply and coupling caps to isolate it's two stages as well as the headphone load.

 

Cheers!

post #6569 of 6797
Quote:
Originally Posted by the_equalizer View Post

For starters, the input stage above looks like differential pair with D1A and D1B acting as a tail CCS; a topology similar to the SOHA II. On the other hand the SSMH is a simple common cathode triode voltage amplifier.  

 

The circuit around Q4 and Q5 is a CCS acting as source load for the output MOSFET stage, not for the tube stage. On the SSMH it is the tube heater that acts as the MOSFET load.

Are you referring to the standard SSMH? In #6561, Goobley brought in the MiniMax (with its CCS tweak) into the discussion . So that's what I was kinda comparing the Borbely to.

 

 

IAC ... your point-by-point contrasts are good. However, I was largely referring to semantics and nomenclature used to generically describe BOTH amps, which is why I used quotations in #6567's remark:

The amps ain't THAT different in that both are "vacuum tube and discrete solid state hybrid headphone amplifier[s]".


Edited by alphaman - 6/21/13 at 3:11am
post #6570 of 6797
Quote:
Originally Posted by alphaman View Post

Are you referring to the standard SSMH? In #6561, Goobley brought in the MiniMax (with its CCS tweak) into the discussion . So that's what I was kinda comparing the Borbely to.

 

 

IAC ... your point-by-point contrasts are good. However, I was largely referring to semantics and nomenclature used to generically describe BOTH amps, which is why I used quotations in #6567's remark:

The amps ain't THAT different in that both are "vacuum tube and discrete solid state hybrid headphone amplifier[s]".

 

The Minimax has also a triode common-cathode voltage amplification stage, like the SSMH, but with CCSs acting as plate loads. The CCS SSMH tweak I did involves adding these plate load CCS to the tube stage. This is of course just what I did, other's could try other kinds of CCS, like buffer op-amps or diodes. Morgan Jones' "Valve Amplifiers" has a very nice section on CCS for use on vacuum tube circuits.

 

As to the Minimax's output stage, it is significantly different to both the SSMH and the Borbley, these last two being single-ended while the Minimax uses a BJT diamond buffer output stage using complementary semiconductor devices.

New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Millett "Starving Student" hybrid amp