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 443

post #6631 of 6757
Quote:
Originally Posted by Tjj226 Angel View Post

Three last questions guys. 

 

http://www.diyforums.org/SSMH/variants/SSMH-12AU7.gif

 

1: As I look at the schematic for the 12au7 mod, I see that there are two 0.22uf caps in parallel with the 470uf caps. The 0.22uf caps are optional, and with such a little change in capacitance, I am wondering if it would be smarter to buy higher quality 470uf caps, or simply add in the 0.22uf capacitors and leave it alone. 

 

2: Resistors R2 and R8 on the schematic are listed as 390K ohm resistors. However, on the BOM page, they are listed as 220K ohm resistors. Which one should I get? 

 

3: Resistors R15 and R14 look like they just essentially lower the volume a bit. Is this true, or do they serve a greater purpose? 

 

1.  They are "supposed" to affect the sound quality, by using high end film caps you can get better sound.  I quote "supposed to" because I don't have golden ears.  I didn't bother using them on mine.  Others argue that bypasses smear the sound.

3.  They prevent oscillation.  Kim's link explains it.  It's the same for the mosfet gate stoppers.  R3/R9.

 

Edit, when you wire it (assuming you don't have the PCB), put the stoppers right on or very close to the device.  I put them right on the socket, and the mosfet pin.


Edited by holland - 8/7/13 at 3:31pm
post #6632 of 6757
Quote:
Originally Posted by Tjj226 Angel View Post

Three last questions guys. 

 

http://www.diyforums.org/SSMH/variants/SSMH-12AU7.gif

 

1: As I look at the schematic for the 12au7 mod, I see that there are two 0.22uf caps in parallel with the 470uf caps. The 0.22uf caps are optional, and with such a little change in capacitance, I am wondering if it would be smarter to buy higher quality 470uf caps, or simply add in the 0.22uf capacitors and leave it alone. 

 

2: Resistors R2 and R8 on the schematic are listed as 390K ohm resistors. However, on the BOM page, they are listed as 220K ohm resistors. Which one should I get? 

 

3: Resistors R15 and R14 look like they just essentially lower the volume a bit. Is this true, or do they serve a greater purpose? 

The last two posts kind of give you an indication that the opinions for capacitor bypassing run the gamut.  It depends on an awful lot of things - including the amp circuit design and the capacitors involved.  You guessed correctly that they're not there for the extra capacitance.  They're there to "bypass" the electrolytic capacitor.

 

Capacitor bypassing is based on a few things that are empirical except #3:

 

1. Film capacitors sound better than electrolytic capacitors

2. Large film capacitors - enough to match an electrolytic in ratings - are so large and so expensive that they're impractical.

3. This is the based-in-engineering part: any audio output capacitor (used for AC-coupling) is going to form an RC circuit with the load (headphones).  This means that depending on the farads and impedances, certain frequencies will pass through better than others.

 

Because of the RC "filtering" effect, it can be argued that the bulk of the very detailed frequencies - highs and midrange - will by and large pass through the small film cap, while the mid-bass and deep-bass will go through the electrolytic.  One might say it's an ingenious way of applying a high-quality film cap for the "most" sound, while the electrolytic ensures that all the bass frequencies still get through.  Some people say there's actually a cross-over of sorts and claim they can hear a "smear."  Others may notice that the highs are clearer, less-distorted and that the bypass arrangement sounds better than the electrolytic alone.

 

One could also argue that the capacitor pair actually forms a brand-new, hybrid-style of capacitor and that the actual frequency bandwidth is the same for both, but the combination itself sounds/performs better.

 

I don't know which is true, but I can tell you that there are definitely cases where the pairing sounds better than an electrolytic along.  There's also very little predictability in whether one pairing sounds better than another.  Generally speaking, Wima film caps (typically MKP10's at 0.1uf, 0.22uf, 0.33uf) perform fairly consistently with many electrolytics.  Even then, though, some combinations work, some don't.  Pairings of other type of capacitors are a crapshoot.  Luckily, there are several sources of documented results with bypass capacitor combinations - Dsavitsk's "Some Notes on Output Coupling Capacitors" at ECP Audio, Humble Homemade HiFi capacitor reviews, etc.

 

Finally, Pete Millett uses Wima's as capacitor bypasses in many of his designs.  We thought it was a nice touch to put on his Starving Student, too.


Edited by tomb - 8/7/13 at 8:21pm
post #6633 of 6757

I wonder one thing. Why C3/C3a/C5/C5a is said to be 63V or even more? There is no more than 12/19VDC on that line. I used 25V capacitors and they're doing the job. Nothing has exploded yet.

post #6634 of 6757
Quote:
Originally Posted by ProTofik View Post

I wonder one thing. Why C3/C3a/C5/C5a is said to be 63V or even more? There is no more than 12/19VDC on that line. I used 25V capacitors and they're doing the job. Nothing has exploded yet.

If I'm not mistaken, if you short a MOSFET (which has happened with a number of builders in this thread), those caps could be exposed to 48V.  Granted, it's a conservative choice, but you usually choose cap ratings on the basis of failure.  It's one thing to short something out or have a bad tube, but it's something else again to blow electrolyte all over the place.  Only C7 and C8 (if you use them) are totally protected - the higher voltage would have to pass through the vacuum in the tubes and that's impossible if they fail.

post #6635 of 6757

Looks like I will have to replace them with 63V ones. Anyway, what are C7 and C8 caps used for? I don't have them in my build.

post #6636 of 6757
Quote:
Originally Posted by ProTofik View Post

Looks like I will have to replace them with 63V ones. Anyway, what are C7 and C8 caps used for? I don't have them in my build.

Those are cathode bypass capacitors.  They're optional, but many people use them in this type of tube circuit.  I don't understand it all myself, but it's explained on pages 75-80 of Morgan Jones' book, "Valve Amplifiers."  The Starving Student circuit uses biasing resistors on the cathode side of the tubes - R5 and R11.  The resistance allows an additional current loop that produces a bit of feedback and reduces gain and available voltage to the tube (I may be stating that incorrectly).  Placing a capacitor across this resistor allows signal frequencies to "bypass" the resistor and pass through the circuit.  This removes the feedback/gain reduction for those frequencies.

 

The cap itself forms an RC circuit with the resistance of the cathode, so you have to size it properly to prevent bass frequencies from being filtered out.  We use 1000uf on the MiniMAX/MOSFET-MAX, for instance, but those are different tubes and trimmers are used as the biasing resistors.  Dsavitsk sized C7 and C8 for the Starving Student PCB, so I'm sure they're sufficient if you want to use them.

 

Truth be told, the gain is so high already on the Starving Student using the 19J6 or the 12AU7 that it may be better to leave them out, but that's up to you.

post #6637 of 6757

It funnels the AC through the cap, allowing the resistor to set DC bias.  It increases gain also (there are equations for it).  Without the cap you get some negative feedback through the resistor, which reduces gain and also increases output impedance of the tube.  You do need to size the cap properly, as tomb indicated, due to RC filtering.  I don't use the cap.  I had it, but removed it in favor of a more usable range on the pot, coupled with global negative feedback.

post #6638 of 6757
Quote:
Originally Posted by tomb View Post

Those are cathode bypass capacitors.  They're optional, but many people use them in this type of tube circuit.  I don't understand it all myself, but it's explained on pages 75-80 of Morgan Jones' book, "Valve Amplifiers."  The Starving Student circuit uses biasing resistors on the cathode side of the tubes - R5 and R11.  The resistance allows an additional current loop that produces a bit of feedback and reduces gain and available voltage to the tube (I may be stating that incorrectly).  Placing a capacitor across this resistor allows signal frequencies to "bypass" the resistor and pass through the circuit.  This removes the feedback/gain reduction for those frequencies.

 

The cap itself forms an RC circuit with the resistance of the cathode, so you have to size it properly to prevent bass frequencies from being filtered out.  We use 1000uf on the MiniMAX/MOSFET-MAX, for instance, but those are different tubes and trimmers are used as the biasing resistors.  Dsavitsk sized C7 and C8 for the Starving Student PCB, so I'm sure they're sufficient if you want to use them.

 

Truth be told, the gain is so high already on the Starving Student using the 19J6 or the 12AU7 that it may be better to leave them out, but that's up to you.

 

Quote:
Originally Posted by holland View Post

It funnels the AC through the cap, allowing the resistor to set DC bias.  It increases gain also (there are equations for it).  Without the cap you get some negative feedback through the resistor, which reduces gain and also increases output impedance of the tube.  You do need to size the cap properly, as tomb indicated, due to RC filtering.  I don't use the cap.  I had it, but removed it in favor of a more usable range on the pot, coupled with global negative feedback.

 

This was really helpful! I had the bypass caps on the cathode without realising that they negated the feedback from the resistor. With the gain on my EW178 so high, I have a lot more usable range on my pot now.

post #6639 of 6757

I got a pair of Grado SR80is last Christmas, and I've been listening to them daily since then. I've always read that they really open up when amped, but I had no idea just how much until I plugged them into my brand-new-build-by-me 12SR7GT Starving Student Millet Hybrid headphone amplifier.

 

Some of the brightness went away, but the detail is amazing and the sound stage really opened up. Some of the music I listen to has a very lo-fi vibe, but on others, like classical and acoustic, the change is stunning. Definitely worth the effort! I think I did pretty good job on wire management, but it's late, so I'll post pictures tomorrow. All in all, I'm very pleased with my first amp build! Can't wait to get some more listening time in.

post #6640 of 6757

Hello Head-Fi!


After reading the 433 pages of this thread over the last couple if weeks, I figured I'd be prepared to build this thing properly with the wealth of knowledge that everyone has provided - apparently not. 


I took things slow during the build, making sure I didn't make any mistakes (~24 hours into this thing), but apparently I didn't go slow enough. When I turned it on, the tubes glowed (19J6 version) and nothing seemed to be going wrong, so I was pretty happy. I didn't have a 1/4 inch adapter that fit with some cheap earbuds, so I plugged in my HD 25-1 II. Big mistake. Right channel is now blown (thankfully still under warranty).

 

I opened the amp back up - one of the output caps had split open.  As far as my knowledge extends, I believe this is the problem. (May have put it in with reverse polarity... 1st timer mistake) From what I understand, this would be supplying the 'phones with too much voltage, right? 


I originally thought this problem was just channel imbalance, but upon playing a left channel/right channel only track, I'm getting a LOT of crosstalk as well. Where in the circuit could this be coming from? Or could it simply be because my output cap was bad?


I originally had the BOM Pot, but I just ordered the Alps pot, so I'm confident that any channel imbalance that I may have had won't be a problem in the future. 

 

I also noticed that I get a massive amount of 'static' when I unplug and plug in my headphones. It only happens during the process of plugging in and unplugging.  I don't think that this is the 'turn on thump' that people are talking about, because it still happens when the amp is left on for more than 5 minutes.  I haven't heard of anyone else having this problem, so I'm interested in finding out what the problem could be.

 

Thanks for any input guys!


Edited by Colin94 - 8/25/13 at 6:23am
post #6641 of 6757

 


Quote:
Originally Posted by Colin94 View Post

I opened the amp back up - one of the output caps had split open.  As far as my knowledge extends, I believe this is the problem. (May have put it in with reverse polarity... 1st timer mistake) From what I understand, this would be supplying the 'phones with too much voltage, right? 

Yes, if the cap fails, it could expose that channel to 19VDC. However, unless you wired it backwards, it's just an unfortunate component failure. I wired mine point to point and it worked fine from the get go, so you never know.

 

Quote:

Originally Posted by Colin94 View Post

I originally thought this problem was just channel imbalance, but upon playing a left channel/right channel only track, I'm getting a LOT of crosstalk as well. Where in the circuit could this be coming from? Or could it simply be because my output cap was bad?

If it was me, I would replace the capacitor before I started trouble shooting other things. My 2 cents.

 

Good luck and keep us updated!

post #6642 of 6757

I would fix the problem before replacing the cap.  You don't need the output cap to debug, pull it out, no need to connect the headphone jack.  The condition that blew the cap exists.  Also, what size cap are you using?

 

Pictures help, before you remove the cap.

 

Always check your build with a DMM.  The final step is plugging in headphones, not the first.

 

Going slow is not really the best method.  Build and test.  You can go as fast as you feel comfortable doing.  I build quick, double check the circuit before applying power, apply power and probe points I feel are relevant (don't short out or you may blow something).  Sometimes I build in probe points if there is no resistor to probe against.  Then, if all is good, I pull out cheap headphones and connect them.

 

Channel imbalance can be caused by a bad circuit, not just the pot.  The incorrect resistors in certain spots will do it.  Bad tubes can do it too.  If you suspect the pot, pull it and hard wire in a resistor divider.

post #6643 of 6757

So, I've got a bunch of extra 19J6's (I think it's 4 sets) at home and I need a new project. 

 

Has anyone seen it used in any other circuits?  I'm thinking preamp or possibly a tube buffer, though it may be a little silly to build something with the funky heater requirement when I could just as easily go with something more common.  All the same, these extra tubes are just gathering dust.

post #6644 of 6757
Quote:
Originally Posted by Sodacose View Post

So, I've got a bunch of extra 19J6's (I think it's 4 sets) at home and I need a new project. 

 

Has anyone seen it used in any other circuits?  I'm thinking preamp or possibly a tube buffer, though it may be a little silly to build something with the funky heater requirement when I could just as easily go with something more common.  All the same, these extra tubes are just gathering dust.

Your best bet would be to sell those to someone who wants them for a Starving Student or build another Starving Student yourself.

 

There's nothing the 19J6 can do that the 6J6 can't do, except for the Starving Student.  Meanwhile, the 6J6 tube numbers in the tens of thousands and the 19J6 ... well, let's just say you could become a 19J6 tube dealer - for a little while, anyway. wink.gif


Edited by tomb - 8/29/13 at 12:11pm
post #6645 of 6757

Tomb,

 

I just bought the iron for a Tubelab Simple SE build, so selling these is probably what I'll do.  I may keep one set, but I use the starving student so infrequently that even those may not ever be used.  I'll put them up in the FS/FT forum at some point soon.  Maybe some one has some KT88's they want to trade :)

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