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 419

post #6271 of 6757
Quote:
Originally Posted by Mshenay View Post

There's some fellow selling a kit on ebay, here's another question

 

Do these have enough power to drive a 250 ohm can? I'm sure the answer depends on who built it and what the used, but following the CHEAPEST build plana [in case I end up with that <.<] how would they do! 

 

It doesn't really depend much on who built it, really, as long as it's built and it operates correctly.

 

Be assured, this amp has power to spare. 250 ohms cans will pose no problem to it. It will also perfectly drive low impedance headphones such as Grado or Denon. I have even used it to drive a couple of small 8 ohm speakers.

 

cheers!


Edited by the_equalizer - 10/30/12 at 8:11am
post #6272 of 6757
Quote:
Originally Posted by kchapdaily View Post

it depends on the can. but millet states on the starving page that his hd600's are well powered with a standard starving student and thats a 300ohm can. ive never used anything high impedance with mine because im an ortho type of guy, but my ssmh poweres my hifiman he-4's well and they are usually difficult to power. generally any amp with tubes will have the required voltage to power high impedance voltage hungry cans.

 

 

Indeed, the HD-600 work very nicely with the "Starving Student". Mr. Millett also mentions he used it to drive his AKG K1000 and was surprised with how good they sounded.

 

Ah! I'm so curious about the HE-X + SSMH pairing. I'm sure this little amp can drive the planar magnetic headphones with enough authority and dynamics.

post #6273 of 6757

it can indeed! my amp is pretty stock, and ive never experienced any clipping or distortion, even at high volumes. the ssmh can get the he-4 to volumes past listenable limits. i think the only hifiman that the ssmh might have difficulty with would be the he-6. 

post #6274 of 6757

AAahh! There was an HE-6 at the last Montreal Meet, but it was way too busy for me to try it on my stuff. Just for kicks, I'll try all the high-end stuff with my SSMH at the next meet. I'm sure I could try the HD800, some Audeze and the HE-6. =D

 

Quote from Millett's original webpage:

The 5% THD point is about 7V RMS into 100 ohms, or 3V into 32 ohms.  If you've ever listened to 3V RMS into a pair of Grados... you've probably suffered long term hearing damage.  For comparison, the Millett hybrid design clips at 2-3V RMS into pretty much any load.

 

I think 7V rms should be sufficient to drive just about any headphones, though. And with a MOSFET at the output, current should not be a problem either.

 

I've actually been curious about building a beefed-up version of the SSMH with 12AX7 tubes and use it to power close-field monitors. Great to see someone else thought about it. I'm not that crazy, after all. biggrin.gif

post #6275 of 6757
Quote:
Originally Posted by KimLaroux View Post

 

I've actually been curious about building a beefed-up version of the SSMH with 12AX7 tubes and use it to power close-field monitors. Great to see someone else thought about it. I'm not that crazy, after all. biggrin.gif

 

 

Gain is not the problem.

 

Stick with the 12au7 for speakers. How many vrms do you put into your nearfeield monitors? Like 2vrms? So you still have a decent bit of voltage headroom. Not an awful lot, but at the same time not an awful little ;) 

 

You can probably find a few more vrms if you go around and realllllly optimize the circuit you have now. Using *just* the heater as the source resistor is a great way to save a few bucks, but not necessarily the best way to get the most voltage swing out the amp. Try to get the source of the mosfet around the middle of the voltage supply +/-a few Vdc. Add a resistor or CCS in series with the heater to make it happen. I would lean towards a resistor if you want absolute maximum voltage swing (despite my love of CCS's) since resistors behave better when you drive them into the rails. CCS's often work better under "ideal" conditions, but if you are really pushing an amp to its absolute limits things quickly become non-ideal. Its something else to experiment with anyways :)

 

You may also want to replace the plate resistors for the tubes with a CCS - this should buy you some basically free gain and improve the linearity of the tube a bit too. 

 

If your interested in more current (have you hit that limit with your monitors? Yea, um no, but since I'm already here) try folding the heater so its 6.3V@300mA rather than 12V@150mA and adding a resistor or CCS to make up for the voltage difference. Ooh, you will need a PSU good for at least 1A and heatsinks like whoa. 

 

Is now the right time to say that the 12ax7 is a sucky tube? Perhaps not the best time, but its never really a bad time to call a sucky tube a sucky tube. In the case of the MHSS or a variant thereof the very high plate impedance will not react well to the input capacitance of the mosfet and unless you have a very low output source you will almost certainly not need the gain after you change the plate resistors to CCS. Just stick with the 12au7, or one of its low gain, low Rp cousins.

post #6276 of 6757
Quote:
Originally Posted by KimLaroux View Post

AAahh! There was an HE-6 at the last Montreal Meet, but it was way too busy for me to try it on my stuff. Just for kicks, I'll try all the high-end stuff with my SSMH at the next meet. I'm sure I could try the HD800, some Audeze and the HE-6. =D

 

I think 7V rms should be sufficient to drive just about any headphones, though. And with a MOSFET at the output, current should not be a problem either.

 

I've actually been curious about building a beefed-up version of the SSMH with 12AX7 tubes and use it to power close-field monitors. Great to see someone else thought about it. I'm not that crazy, after all. biggrin.gif

EDIT: Right after posting I saw Nikongod beat me to it :)

 


I used it to drive a couple of small speakers, like the ones in old transistor radios. Of course the sound wasn't great but it was certainly listenable. While using a 12AX7 will get you more voltage gain from that first stage (even more if you use CCS loads), given the very low plate voltage I don't think it'll get you more headroom. From what I have read you need to throw around 200 volts at the plate of a 12AX7 to bias it out of grid current.

 

Now, of course, I would definitely try it :) I think your build uses the 12AU7, so you just need to plug in a couple 12AX7 and hook up a couple monitors to the output; no need to change any biasing resistors.

 

Another idea is to build a tube gain stage with "proper" plate voltages and the same MOSFET output stage. It would require two power supplies. You could still use the tube heaters as MOSFET loads. Yet from there it should be easy to add another 6.3 volt supply for the heaters and use a current source to load the MOSFETs. Use a couple other more CCS to load the tubes. Or maybe, while you're at it and since we already have plate and heater supplies, throw away the 12A_7 and use a couple 6DJ8, maybe in cascode configuration. Finally make the output stage power supply bipolar and do away with the output cap by using complementary MOSFETs in a class-A push-pull configuration. Something similar to this.  Or maybe take one of Nelson Pass "Zen" output stages, and use a cascoded 6DJ8 as input stage... Now I only need a couple of months and maybe $1,500.00 USD in parts to build my dream  :).


Edited by the_equalizer - 10/31/12 at 9:57am
post #6277 of 6757

I use the low-power headphone amp (there was an optional hi-power version) in a Zhalou to drive my single driver speakers. They fill a medium sized room very nicely. :) 

 

Efficient speakers are very fun with powerful headphone amps. 


Edited by nikongod - 10/31/12 at 10:05am
post #6278 of 6757

mmm the more I hear the more I want it :D 

 

What's the input options? RCA in would be nice! 

post #6279 of 6757

You build it yourself.  Lots of people have built it with RCA inputs.

post #6280 of 6757
Quote:
Originally Posted by Mshenay View Post

mmm the more I hear the more I want it :D 

 

What's the input options? RCA in would be nice! 

 

Quote:
Originally Posted by Lou Erickson View Post

You build it yourself.  Lots of people have built it with RCA inputs.


Indeed, a builder in this thread even built it with switchable RCA  *and*  USB inputs (a Bantam DAC, if I remember correctly, built into the same enclosure)

post #6281 of 6757

Neat-o, I might want the RCA for simplicity. Seeing as teh Dac I'm gunning for has RCA out and I'm getting some Audio Quest RCA interconnector cables

post #6282 of 6757

I just finished my amp and now its onto the debugging stage. 

 

1 I turned it on, no smoke good sign 

2 felt heat sinks, only the left channel warmed up, I quickly spotted that I hadn't connected the other mosfet to the power

3 voltages, I have 16 volts at the source of the mosfets? I should only have 12 any idea what i could have done/mis-wired to make it this high?

 

EDIT  one is at 14.8 the other is at 15.8 , is this okay or do i need to change something?

EDIT 2 i went ahead and plugged in some cheapo ear buds, i get sound but alot of hum and a clicking sound about every 2 seconds.


Edited by shamazo - 11/4/12 at 9:14pm
post #6283 of 6757

The clicking noise is probably wireless interference. Try turning stuff off until you find the source: cellphone, laptop, router... The humming may be internal or external. Try moving the amp around while it's on. Mine stops humming if I move it away from my DAC.

 

16 volts is normal. Keep in mind that the tube heaters are simply resistive elements. They are rated for 150ma at 12v, but since the mosfet pushes more than 150ma into them, you will see a higher voltage across. This current is set by the bias of the mosfets, R2/R3 and R8/R9. Higher voltage at the mosfet gate means more current trough the tube heater, means higher voltage drop across the heater. Check the voltages between the gate and the source (Vgs) of each mosfet. It should be somewhere around 4v. Since you have more voltage across one tube, this one should have higher Vgs. It's probably unbalanced because you did not match the resistors, and their values are far apart. You could always rebuild the bias networks, so the voltages are closer together. Keep in mind that the mosfet will turn off at 2V, and that you need some headroom. 4v is a safe margin. Too much and you'll blow the tubes, or you'll overload the PSU. Wait a couple of minutes for things to warm up before taking measurements.

post #6284 of 6757
Quote:
Originally Posted by KimLaroux View Post

The clicking noise is probably wireless interference. Try turning stuff off until you find the source: cellphone, laptop, router... The humming may be internal or external. Try moving the amp around while it's on. Mine stops humming if I move it away from my DAC.

 

16 volts is normal. Keep in mind that the tube heaters are simply resistive elements. They are rated for 150ma at 12v, but since the mosfet pushes more than 150ma into them, you will see a higher voltage across. This current is set by the bias of the mosfets, R2/R3 and R8/R9. Higher voltage at the mosfet gate means more current trough the tube heater, means higher voltage drop across the heater. Check the voltages between the gate and the source (Vgs) of each mosfet. It should be somewhere around 4v. Since you have more voltage across one tube, this one should have higher Vgs. It's probably unbalanced because you did not match the resistors, and their values are far apart. You could always rebuild the bias networks, so the voltages are closer together. Keep in mind that the mosfet will turn off at 2V, and that you need some headroom. 4v is a safe margin. Too much and you'll blow the tubes, or you'll overload the PSU. Wait a couple of minutes for things to warm up before taking measurements.

thanks, ill do some measurements soon. would the interference just be in the tubes or through the whole circuit? as my tubes are in sockets that cover them in aluminium.

also roughly how long to let it warm up? is 5 minutes enough?

 

EDIT i have 3.5 and 3.65 after about two minutes.

EDIT 2 i get a large increase in hum when i touch the volume pot. it also makes a kind of scratching noise when i move it, but i don't really mind that. I think the interference could be a mixture if internal and external, it increases when i move the power supply closer, but doesn't entirely go away when i move it even further away. even if i surround the entire thing with tinfoil. I'm guessing its a ground loop? any ideas on finding the source/fixing it?

Probably not a problem, but are the tubes meant to be hot to the touch?


Edited by shamazo - 11/5/12 at 1:55am
post #6285 of 6757

Yeah tubes do get hot, it's how they work. 5 minutes is enough to take measurements.

 

So you're actually using tube shields, but you still get noise? Meh. I too have been fighting with all sort of noises out of this amplifier. I bought a couple of shields for the tubes, but haven't had the chance to try them yet. The Faraday cage I originally built is good to block hum, but not wireless. The tubes themselves will pickup interference and amplify it, but the circuit around the tubes is also sensible to EMI. Make sure that no power wire come close to the tube sockets. Moving wires around may help. The input is a high impedance circuit and is also very sensible to interference. In my build I get hum if I just move a finger or a metal tool close to it. The input wire should be kept as far away as possible to the power wires, and if they need to cross make sure they do it at a right angle.

 

As for the pot buzz, you have to ground the pot's shaft. This fixed it for me.

 

Is your enclosure built of metal? My original build was plastic, and it was unusable. It picked up just about every single stray EMI in my room and amplified it. I rebuilt it in aluminum, which helped but isn't perfect. My next one will be in a ferrous enclosure. Though with the number of people these days who build it and give up because it's too noisy, I might think about it twice if I ever use the design again.

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