[Dan55]

Thanks very much for the response, tomb. I'm measuring 27V for the power supply (V+ to ground), 13.5V for each tube bias (TA2L/R to ground), and 75mV for each DB bias (TB1L/R to TA2L/R). I think the problem is probably in the signal path somewhere, particularly the 1/8'' input jack I'm using that is connected to the back of the board. The initial buzzing I was getting was affected when I moved the wires from this jack to the board around a bit.  If I plug an interconnect into that jack, though, I'm still measuring continuity between the interconnect and the input pads at the front of the board, next to the pot. I'm not sure if this indicates that the input jack is okay.
 
The output jacks are the only part of the amp that have seen heavy usage. I have not replaced the tubes and the source remains connected to the input jack.
 
Sorry for the lack of specificity. I'm a bit stumped and posted hastily yesterday when I lost all output and had to start using my computer's audio jack! Please let me know if there are anymore measurements I can take to narrow down the problem. 

 

[tomb]

Quote:

Thanks very much for the response, tomb. I'm measuring 27V for the power supply (V+ to ground), 13.5V for each tube bias (TA2L/R to ground), and 75mV for each DB bias (TB1L/R to TA2L/R). I think the problem is probably in the signal path somewhere, particularly the 1/8'' input jack I'm using that is connected to the back of the board. The initial buzzing I was getting was affected when I moved the wires from this jack to the board around a bit.  If I plug an interconnect into that jack, though, I'm still measuring continuity between the interconnect and the input pads at the front of the board, next to the pot. I'm not sure if this indicates that the input jack is okay.
 
The output jacks are the only part of the amp that have seen heavy usage. I have not replaced the tubes and the source remains connected to the input jack.
 
Sorry for the lack of specificity. I'm a bit stumped and posted hastily yesterday when I lost all output and had to start using my computer's audio jack! Please let me know if there are anymore measurements I can take to narrow down the problem. 

OK - it sounds like there's an issue with this 1/8" jack you're using.  How about getting rid of that and connect some RCA jacks with new wire into the input terminal block?  If your only source connector is through an 1/8" mini-jack, then get a convertor patch cord.  It's pretty easy to find mini-to-RCA convertor patch cords at Walmart and other places.  Doing this would at least confirm there's nothing wrong with the MAX.
 
 

 

[Dan55]

Thanks so much for the help, tomb. The problem is indeed with the output on my DAC. I was using a second output on the DAC for another amp and it didn't occur to me that the output jack I was using on the DAC was just as likely to have gone as the input jack on the amp. I'll try troubleshooting the DAC for a while before bothering AMB with my newb questions.
 
The amp remains flawless and sounds even better after having had to listen to my computer's audio for a few nights. Thanks again!

 

[ryssen]

I wonder what the sound difference beetwen mosfet and Bjt´s are.Anyone kompared the 2?

 

[tomb]

 
Quote:

I wonder what the sound difference beetwen mosfet and Bjt´s are.Anyone kompared the 2?

Well, sort of obviously ... I have built both and listened to both.  Fundamentally, with a high-impedance phone, the MiniMAX (or BJT MAX) is superior and probably has more headroom.  However, with low-impedance phones or small speakers, the MOSFET-MAX is superior.  I would define low-impedance as 32 ohm Grados and high-impedance as 300 ohm Sennheisers.  Between those two limits, it gets much more difficult to tell the difference.  Also, both will do pretty d*mn good with the other's optimum.  IOW, a MiniMAX (or BJT MAX) will still do pretty good with Grados, while a MOSFET-MAX will do OK with Sennheisers.  It's just that they're not necessarily optimum at that.
 
The MiniMAX can deliver 50ma in Class A to a low-impedance load.  At the same time, the MOSFET-MAX can swing as much as 11.5V in either direction for a high-impedance load.  This is in the opposite direction of the optimums for both amps, yet both of these values exceed the limits of many solid-state amps (especially portables).
 
The MOSFET-MAX is probably less successful with high-impedance than the MiniMAX is with low-impedance.  But there's not many things that can drive to low-impedance headphones with as much power and detail as a MOSFET-MAX.  Other MOSFET-output DIY headphone amps can deliver more power, perhaps. but not with a tube front-end, AFAIK.  The MOSFET-MAX has the potential of delivering almost 1.25W, if necessary, with certain settings.
 

 

[ryssen]

Thanks for the explantion,I will then build a BJT as I have the Sennheizer HD650.

 

[chadbang]

I'm just wondering if I'm imagining things (I don't think so) but I use my Millet Maxed (BJT) and work. It sounds okay for the first three hours of so, and then it sort of seems to "bloom" with bass and sounds warmer and fantastic. I've noticed this every day for two weeks. Anyone else have a similar experience. Is this amp just very slow to warm up? Cause when it does...wow. i'm in heaven.

 

[tomb]

Quote:

I'm just wondering if I'm imagining things (I don't think so) but I use my Millet Maxed (BJT) and work. It sounds okay for the first three hours of so, and then it sort of seems to "bloom" with bass and sounds warmer and fantastic. I've noticed this every day for two weeks. Anyone else have a similar experience. Is this amp just very slow to warm up? Cause when it does...wow. i'm in heaven.

A couple of things -
1. Yes, for some reason - maybe that they're low-voltage tubes - the Millett-type tubes can take a long time to warm up for the best sound.  That said, I would say the peak is ordinarily an hour, not 3.  However, some tubes may vary longer than that, it all depends.
2. You might check your tube bias during the period you describe - maybe once an hour.  You may find that the tubes "float" a bit in bias over a few hours and perhaps there's some sort of optimum bias point that results in the sound you enjoy.  I've seen some tubes continue to drop their bias by a few tenths of a volt for several hours.

 
 

 

[ryssen]

As I am building a Millet Max BJT I wonder how tall the heatsinks should be,and how many mA I should bias it to?

 

[tomb]

Quote:

As I am building a Millet Max BJT I wonder how tall the heatsinks should be,and how many mA I should bias it to?

Normal method with BJT's is to bias to 50ma (110mV on the test points) and use 1" tall heat sinks.  This is what we do on the MiniMAX.  However, I've had reports from one builder who claimed to notice a difference when biasing the BJT's to even higher currents on the regular MAX.  If you use 1-1/2" tall heat sinks, you could bias all the way up to 125ma, which is the recommended bias for a MOSFET-MAX.
 
The MAX (as opposed to the MiniMAX) uses the Hammond 1455T1601 as the recommended case.  It will accomodate 1-1/2" heat sinks if you want to use them.  Don't forget to make the power supply heat sink 1-1/2", too.  It will draw much more current - in fact, there's a bias point where the heat of the power supply sink actually takes over as the hottest part on the PCB.  If you bias up that high, you also need to use a 1000ma walwart (WAU24-1000).  We depend on the higher current rating to give us those extra few volts to 27VDC at the V+ point on the PCB.  If you stay with the lower current rating walwart, there will be enough current most likely, but you'll lose a couple of volts to stay in regulation.  (When adjusting the power supply trimmer, notice the voltage point where you can't adjust any higher, then back off by at least 1 to 1.5V.  That's your maximum voltage to keep the power supply in regulation, as in low noise.  Typically, you should be able to get to 28-28.5VDC, then back off to 27VDC as the final setting.  If you turn the bias current up past 50ma and don't use a higher current walwart, you won't be able to get this much voltage.)
 
Lastly, remember that BJT's will drive into thermal runaway if you're not careful.  They aren't as tolerant as MOSFETs at high currents and temperatures.  Those biases above should work with the taller heat sinks, but you need to be very careful that you don't push them higher while you're adjusting/setting things.  What will happen in thermal runaway is that as the BJT's heat up, they'll draw more current on their own.  If they draw more current, then they heat up even more, which makes them draw more current ... and if this continues, things get nasty pretty quick.

  Also remember to check the bias  with the case lid on (using more tip jacks on the back or some other method). BJT's will run hotter closed up (MOSFETs do the opposite).  So - always keep an eye on things, wait on the BJT's to heat up each time with small increases in bias current, and GO SLOW.
 
Good luck!
 
 

 

[ryssen]

Is seems hard (Impossible) to find a WAU24-1000 för 220V that I have.I am going to use a 30VA torid,if that is allright? (I have to use a bigger case)

 

[tomb]

Quote:

Is seems hard (Impossible) to find a WAU24-1000 för 220V that I have.I am going to use a 30VA torid,if that is allright? (I have to use a bigger case)

Oops.  I didn't realize you were outside the US.  Sure - 30VA at 24VAC should be more than enough.
 
 

 

[funch]

Has anyone here tried the MOSFET Max with Thunderpants?

 

[tomb]

Quote:

Has anyone here tried the MOSFET Max with Thunderpants?

Not specifically with the actual Thunderpants, but remember cetoole designed it (the MOSFET-MAX).  He let me audition a couple of orthos with it (modded Yamaha's and modded older/retro Fostex) - not to mention that I owned the TR50rp's myself until Smeggy talked me into selling them to him (availability was low at the time and he needed the raw materials for his thunderpants).  The thing to remember is that depending on the tube and the bias setpoint (which can vary), you can swing up to 12V plus or minus at a minimum of 125ma (1.5W) - and that's just in Class A.
 
 

 

[scootsit]

Quick question. I'm just finishing up my Mosfet Max, and am starting on the casework. The diagrams are a little confusing to me. Should I be drilling out the top and bottom of the case or just the top? It would make sense that it be both so that air flows through the case, but hey, what do I know? Anyway, any help would be appreciated!
Thanks!