[the_equalizer]

Cool case! Reminds me of the old tube radios I used to restore 15 years ago. How are you liking the amplifier ?
 
cheers!

 

[bengoshi2000]

Thanks! I picked this case for that exact reason... the retro radio look (even though it's just an old antenna rotor contol box).

I'm REALLY liking the amp. It drives my old AKG240 Monitors ridiculously well. I'd forgotten how much I like those cans!!

I have noticed that the amp *can* reveal some of the deficiencies of lower bitrate mp3's, specifically in the highs. I've ripped my cd's at 192kbps and those sound spectacular... I really think the tubes work well to hide artifacts at higher bitrates. I've got a couple albums from iTunes that I'm now disappointed in as it's painfully obvious that they were ripped at a lower AAC rate.

The amp really does blow my mind. I can pump any type of music through it and it just shines. Daft Punk sounds like I'm driving their mixing console. I can hear each level of layered synth effects with clarity that makes my eyes water. Classical music sounds like I'm sitting in the concert hall live. Strings have a vibrance and clarity that i dont think ive ever heard before. The soundstage is expansive! AC/DC makes me feel as if I've been possessed by Angus Young!

My only complaint is now I have a new mental health diagnosis... MTAAD (Multiple Tube Amp Acquisition Disorder).

...but I'm not complaining...

 

[baintronic]

im sure its in here somewhere but what is the mouser link for the "star" heat sinks everyone is using?
 
thanks!
 
 

 

[tomb]

Quote:

im sure its in here somewhere but what is the mouser link for the "star" heat sinks everyone is using?
 
thanks!
 
 

If you're talking about the extruded, finned heat sinks - this one is 1-1/2" tall and is about as big as you can get and still be "standard" for TO-220 devices:
[size=10pt]647-15ABP[/size]
 
We use it as the power supply heat sink on the MOSFET-MAX and a 1" tall one is used on the MiniMAX.  It's got pins for PCB's, but if you poke around a bit on the Mouser catalog page link, you should be able to find one without so that you can thread it for a point-to-point build.  Or, you can do like Pete did on the original and solder washers on the bottom of the pins to hold them to the top of your case (see first post in this thread).
 
The SSMH PCB kit used a different heat sink altogether - a stamped, bent aluminum affair that was actually for use with TO-3 transistors.  That one is in the BOM on the SSMH website if you're interested in that one.

 
 

 

[baintronic]

Thanks for the rapid response!
 
I ordered those up this morning.
 
Another thing i noticed is that the BOM links to mouser send you to 1/8 resistors .... which i purchased.....
 
is this going to cause a huge problem?
 
thanks
 
 

 

[tomb]

Quote:

Thanks for the rapid response!
 
I ordered those up this morning.
 
Another thing i noticed is that the BOM links to mouser send you to 1/8 resistors .... which i purchased.....
 
is this going to cause a huge problem?
 
thanks
 
 

It's military-spec vs. commercial spec.  Mil-spec is de-rated by one-half.  So an RN-series (mil-spec) V-D resistor is listed as 1/8W, but its commercial rating is 1/4W.
 
 

 

[Starfire]

Looks nice, keeping that back open for cooling I hope.

Good job!

 

[SpringHalo]

I've returned from a lul due to lack of freetime with a completeted starving student amp! Shortly after the arrival of my shipment of parts, I got to work assembling and a mere two days later I had a functioning amplifier!

The front and back panels are made from 1/8" aluminum, and the top and bottom portions are beech. I still have to get a lacquer finish on to tidy up the appearance. I had a very generous friend that let me use his milling machine to make the front and back panels, as well as transform the heatsinks from one large sink to two smaller ones with tapped holes in them for easier assembly. The silver switch on the back is a SPDT brightness switch with Low, high, and off positions for the tube LEDs. I haven't gotten a chance to pick up some RCA plugs for the input, so I've got a temporary 1/8" female plug tethered on to the back.

Enough of the babble, let's get on to the pictures!
 

 
Here's a front shot, with LEDs on 'high' brightness. I really wish that 12AT7s had more open heaters so I wouldn't have to use external lighting, but I'm satisfied with the way they turned out.
 

 
Backside shot, showing a glimpse of my altoids-tinned AlienDAC.
 

 
And one final shot next to my recently acquired DT880s -- they sound great together!
 
I'd like to give thanks to my friend for the tubes and milling machine usage, my father for providing some of his precious beech, and everyone who helped me out with any questions I had in this thread. Thanks!
 
-Spring
 
 
 
On a side note, there's a hiss that is only apparent when there's no music playing and the volume is at ~75% and higher, I think it might be a grounding issue, and there's not much to do to fix that.

 

[bengoshi2000]

Very nice looking build Spring, very nice indeed!  Did you build as a point2point or use a strip board of some sort?
 
With regards to the hiss are you sure it's the amp and not the source?  

 

[the_equalizer]

Slick looking build, indeed. Congratulations!

 

[SpringHalo]

Quote:

Very nice looking build Spring, very nice indeed!  Did you build as a point2point or use a strip board of some sort?
 
With regards to the hiss are you sure it's the amp and not the source?  

 
It's a strict point-to-point build with a sticky copper foil ground plane. Another notable feature I forgot to mention earlier is a switch included in the potentiometer that isolates the ground of the headphone output so you can turn the amp on and off while having headphones still plugged in. I'm almost positive that it's not the source, as the source is an alienDAC that when plugged into a cMoyBB has no hiss or static. I also noticed that the hiss was a lot worse when no source was connected, and that the hiss decreased significantly (about half or so) when I touched any ground part of the build. This was only apparent when I had some cheapo low impedance cans plugged in. With my 600 ohm beyers, the hiss is only apparent when the volume is up to ~60%+ when there's a source plugged in. Comfortable listening volume is around 20-50 though, so it shouldn't be much of a problem.
 
 
 

Slick looking build, indeed. Congratulations!

 
Thanks! I spent a lot of time on it and I'm already planning my next DIY project! There is one other tiny problem that I have -- my original plans called for the tube holes to be 2 inches from the front and 1.5 inches from the sides, but I got them flipped, so now my proposed addition of metal rings around the tubes is messed up. Luckily, I've got an extra bit of beech with the same dimensions, so look forward to version 1.5!

 

 

[KimLaroux]

I came upon this thread while going trough the DIY forum. I'm really impressed by the simplicity of this design. Using the tube's heaters in place of resistors to bias the output mosfet is just brilliant. 

 
Conveniently enough, my room mate gave me 3 GT-12AX7 marketed for fenders. It was used in his guitar amplifier. I used the search tool for 12ax7 and saw that it can be used with the 12a_7 variant. Not much talk on the final result though. From what I can read in the datasheets, the 12ax7 has an "amplification factor" of 100, while the 12AU7 has one of 20. Does it mean the 12AX7 has 5 times the gain of the 12AU7? If that's so, I doubt the 12AX7 I have could be used for this design. Anyone can explain to me what the "amplification factor" is and if the 12AX7 can be used with the SSMH? 
 
Quote from http://en.wikipedia.org/wiki/12AX7:

 Typically a 12AX7 triode is configured with a high-value plate resistor, 100k ohms in most guitar amps and 220k ohms or more in high-fidelity equipment. Grid bias is most often provided by a cathode resistor. If the cathode resistor is unbypassed, negative feedback is introduced and each half of a 12AX7 provides a typical voltage gain of about 30; the amplification factor is basically twice the maximum stage gain, as the plate impedance must be matched. Thus half the voltage is across the tube at rest, half across the load resistor. The cathode resistor can be bypassed to reduce or eliminate AC negative feedback and thereby increase gain; maximum gain is about 60 times.

 
There's a few things here that I'd like to get educated on. First, the plate resistor on the SSMH are 33k, that seems like a long shot from what they say on WP. Is this because the SSMH was designed to drive headphones, or is it that much different for the 12AX7? Then there's the bypassing of the cathode resistor. I read that a few times. What's that exactly? Is it adding a capacitor in parallel to the resistor? If yes, then it seems like it's already added to the 12AU7 variant of the design, is that correct? If yes, then I conclude that removing it will cut the gain in half, making it possible to use 12AX7 on the SSMH. Am I wrong?
 
Though I'd really like some feedback on the difference between say the 12AX7 and the 12AU7. Does one sound better than the other? Is it worth the trouble to modify the design for my used 12AX7 or should I just order new 12AU7?

 

[SpringHalo]

(in regards to C7 and C8)
 
Quote:

Originally Posted by the_equalizer /img/forum/go_quote.gif
 

Yes, that is basically all they do, you can remove them with no ill effects.
 
If you want a bit more detail...  some people prefer to have them since they do away with the local negative feedback generated by the cathode resistor (negative feedback is taboo to many audiophiles) and stabilize the quiescent point of each triode; others dislike them for the reason that they are in the signal path (capacitors, particularly electrolytics, in the signal path are taboo to many audiophiles); even further others dislike cathode resistor biasing for these previous reasons and use other means of biasing the tube.   In synthesis, it's pretty much a matter of personal preference if you want them there or not

 
 
cheers!

 
I used 12AT7 tubes in mine, which have a gain of 60. They run my DT880/600s quite well, but the gain was a bit high (comfortable listening volume at ~90 degree pot turn) so I disconnected C7 and C8 (which were in parallel with the cathode resistor), and now the comfortable listening volume is around half volume. I'm using 100K resistors for R16 and R17 and that seems to work (had 50K but changed that too). If you've got efficient headphones (less than 250ohm I'd assume) I would go ahead and buy the 12AU7s. If not, the only noticeable difference between caps and no caps is a slightly louder buzz when the potentiometer is turned all the way down (which you doesn't really matter). It irks me a little bit, so I might put the Caps back in but you might have a different opinion. Heck, if you get both tubes you can try both and see what you like best.

 

[the_equalizer]

@KimLaroux  This being a build thread, discussion of sonic details is not encouraged, though certainly not forbidden; just to keep things sane and oriented to the 'build the amp' side of things.
 
I did try a couple of nice Mullard 12AX7 tubes with enclosed heaters when I built the 12AU7 prototype. As expected from their higher amplification factor, they produce quite more gain than the 12AU7. I don't think one sounds better than the other, just different. In any case, you can build the amp with your current tubes and, if you're curious you can order 12AU7s (or 12AT7, 12AV7, etc.) afterwards and tube roll 'till you drop .  Beware that if you're planning to use the amp with very sensitive 'phones, IEMs for instance, the 12AX7 will be problematic as the gain they produce will be hard to control. If you're planning to use "stiffer" headphones, you won't have much trouble. I tested mine with AKG K701's and had no problem controlling the volume.
 
You don't *need* to modify anything from the 12AU7 schematic to use 12AX7s: just plug them in. As SpringHalo mentioned in his post, he used 12AT7s in the same circuit. I'm sure you can certainly optimize the circuit for these other tubes by changing the plate resistor and/or the cathode resistor but I suggest you first build the amp as is and, once you get it working, play around with component values.
 
The reason the plate resistor value you're seeing is so different is that that's the value chosen by Mr. Millett for the original design using 19J6 tubes. When I built the 12AU7 version I simply reused my 19J6 with it's 33K plate resistors. I found the 12AU7 did not produce any evident (to my ears) distortion so I did not see a need to find a better value for those resistors.
 
Remember also that the typical values you found in Wikipedia (and that you'll find in most tube manuals) are for high voltage supplies, like those in guitar amps; usually in the 200 volt range. The SSMH uses a 48 volt supply. At such low voltages, the behavior of the tube changes quite a bit so that typical values for biasing resistors are not very reliable.
 
RE cathode bypass caps: Indeed, the 12AU7 schematic includes cathode bypass capacitors, marked as optional. Additionaly to what SpringHalo posted above... you are indeed correct, removing the cathode bypass capacitors will reduce the gain of the circuit (though not as much as half IIRC), so it can help tame the high gain that the 12AX7s produce in the circuit.
 
Finally, the amplification factor of a tube  (simbolyzed with the greek letter 'mu') does not translate plainly to the gain the tube will produce in a given circuit but it does give you an aproximate idea of how much gain a tube will produced compared to others. If I recall correctly from my electronics theory books, it is the voltage gain the tube would yield in a *perfect* circuit, that is, one with no resistive loss, phase shifts, impedance changing with frequency and all those other constraints that real life circuits (as opposed to ideal circuits) have to deal with.
 
Good luck and be sure to post pics of your build here!
 
cheers!

 

[KimLaroux]

Thanks for the complete and detailed answers, The_Equalizer. I didn't mean to ask about the difference in tubes as a discussion to find the best options, but simply to know if the one I have is a compatible part I can build the amplifier around.
 
Good to know the 12AU7 variation circuit can work with the 12AX7 without modifications. I'll do as you recommend then : I'll build the circuit as it is and try it out with the 12AX7. I'll see then how things goes. I have the AKG Q701, so it might be that gain won't be a problem. 

 
In the spirit of the Starving Student, I'll try to build the amplifier using as many recycled parts as I can. So it might take a while to build it. I'm not in a hurry anyways. I'm thinking about building it inside an old car stereo.. might look nice. And I could probably recycle the knobs and pots.