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 215

post #3211 of 6115
5/21/09 at 10:42am
Well, now I've gone and done it. After taking my time and wiring up the power supply/switch/led/pot very nicely, I had a lot of coffee and decided that I should wire up the 'right' channel and give it a whirl.

I guess I rushed...when I turned it on, there were some loud audible bursts of noise (not through the headphones which weren't plugged in, just audible in general) which coincided with the LED shutting off. I quickly took the bottom of the case off, thinking that maybe something had shorted out and sparked inside of it, and the audible click/bursts stopped, but the tube never 'heated' up. I verified that I still had a solid 48v coming in from the back, but no 19v coming out of pin3.

I'll post big ugly detailed pictures after work today, but I was wondering if anyone with more experience than me had some intuition regarding what I most likely did...did I destroy the mosfet maybe?
Reply
post #3212 of 6115
5/21/09 at 3:11pm
Ack! That's not good
Reply
post #3213 of 6115
5/21/09 at 3:41pm
My plan is to just reverse what I did last night ... I'll probably have to re-buy about $5-10 worth of parts (no big deal, but glad I only did one channel).

For now, I still have an untouched, unfinished, hopefully un-broken left channel I can try out. I'll scan in and upload my planned wiring diagram and hopefully someone can slap me upside the head and point out an obvious error
Reply
post #3214 of 6115
5/21/09 at 5:53pm
Quote:
Originally Posted by tomb View Post
Actually, I'd get them matched if you are going this route. Use of this tube series may work fine with the suggested mods, but those are only single-triode tubes. In the 19J6, there are dual triodes that are paralleled together. This tends to average out the differences in tubes and minimizes the need for matching.

With single triodes, the differences in matching from tube to tube may be quite stark, IMHO. No offense to the excellent work finding a working replacement, but I think this is one of the reasons why Dsavitsk suggested that a 12AU7 modification might work a lot better (with similar mods) for the SSMH than this series of tubes.
A double triode may help with differences in triode sections, but that also depends on the relative tolerances. If the tolerances of one single triode tube is very tight, then it may very well be that two of those tubes are more closely matched than two double triode tubes that have looser tolerances.

Not knowing the tolerances, matching would be ideal. However, at $1, you can get a few tubes and if two don't sound right, you can try changing them. Tube matching doesn't always guarantee they work the same in an amp anyways.

I chose this tube because it was very inexpensive. The millet hybrid minimax uses single triodes, and if I recall, beezar used to sell them without matching.
Reply
post #3215 of 6115
5/22/09 at 3:25pm
Quote:
Originally Posted by tomb View Post
I had input resistors, too. Although, I think we were suspecting the stuttering on startup as an incorrect diagnosis. This time I fried the Bantam when the SSMH (or MHSS) had been on for quite awhile. I disconnected the RCA jacks from the SSMH and immediately got a lot of hum. I should've known something was wrong at that point, but just thought perhaps the lack of signal ground caused it. In trying to connect the Bantam, I clumsily touched the center pin of the RCA patch cable to the outer surface of the jack on the SSMH. That's all it took - that channel is dead now on the Bantam.

A few of us suspect that the Cisco's power supply ground "floats" from time-to-time. Unfortunately, it's intermittent and unpredictable. I once measured as much as 0.49V offset between the ground on the SSMH and a capacitively-coupled ground on the Bantam (the dead one, now - OK for testing ). Then it went away and I couldn't measure it again. If this is what is happening, that's why an Alien or Bantam DAC would get fried - the grounds/output on those DACs are not isolated from the amp.

I don't have a scope, but it would be interesting if someone conducted some similar tests on the Cisco power supply ground, just to be sure.

P.S. One other line of suspicion is a shorted tube. Those can easily be intermittent. I think Nate recently said his PCB prototype had experienced a lot of hum lately - it turned out that one of his tubes developed a short. Knowing that the cathode is directly grounded through the MOSFETs and the rest of the amp, this might make a lot of sense, too. Someone correct me if I'm wrong, but if the tube has an intermittent short, it may sound fine most of the time.
interesting... thanks for the response Tom. I'll definitely count my blessings then since my Bantam and MHSS(SSMH? dunno correct acronym hehe) have been operating each weekday flawlessly. Listening to them with the 580s right now

One difference though, is my cable BantamDAC uses a mini phone plug (3.5mm 1/8") as the output. It's a cheaper connector, but maybe the TRS design is more resistant then RCAs?

I'm just gonna continue my practice of unplugging the DAC whenever I turn the amp on. I haven't been doing that when I turn it off. You have me wondering if maybe I should?
Reply
post #3216 of 6115
5/22/09 at 3:50pm
Ok I give up,

I have all of the things that I need to make this amp, and most of all, the time. ( Finally Graduated )

I have tried to mount one of the heatsinks by removing the pins and inserting some screws. Since it is aluminum it takes steel screws pretty well. Not this heatsink. It ate one of my screws, so I drilled it out and went to re-tap it and the damn thing ate my tap.

Now I am at a loss.

How does you mount your heatsinks to the chassis?

I am going to run to ratshack to find another one.

Also a little while back I was able to get two tubes, one a raytheon and the other a tung-sol. Both have the halo's. Will I have any problems?
Reply
post #3217 of 6115
5/22/09 at 3:53pm
Double post......Seems to happen alot lately?
Reply
post #3218 of 6115
5/22/09 at 11:23pm
Quote:
Originally Posted by Logistic View Post
<snip>The millet hybrid minimax uses single triodes, and if I recall, beezar used to sell them without matching.
The Millett Hybrid/MAX/MiniMAX family of amps have used trimmers to specifically adjust bias on the tubes since day one - all the way back to Pete's original Audio Xpress article in 2001. The adjustable bias allows one to always "center" the stereo signal. It's true that if the tubes aren't perfectly matched that there will be differences in dynamics from one channel to the next, but it usually provided good enough compensation until we refined the MAX to the point that greater precision was needed, and started offering matching. After all, these days - there's a lot of less expensive competition nipping at the MAX's heels.

P.S. YGPM.
Reply
post #3219 of 6115
5/22/09 at 11:39pm
Quote:
Originally Posted by Listen2this1 View Post
Ok I give up,

I have all of the things that I need to make this amp, and most of all, the time. ( Finally Graduated )

I have tried to mount one of the heatsinks by removing the pins and inserting some screws. Since it is aluminum it takes steel screws pretty well. Not this heatsink. It ate one of my screws, so I drilled it out and went to re-tap it and the damn thing ate my tap.

Now I am at a loss.

How does you mount your heatsinks to the chassis?

I am going to run to ratshack to find another one.
I'm one of those swimming against the stream, I guess - I don't like threaded and tapped heat sinks. Actually, the pins should work quite well if you use a copper ground plate as Pete used in his original. Look carefully at his pics and you'll see he drilled the holes for the pins, pushed the heat sink pins through, and soldered washers on the back side to the pins and to the plate. Looks pretty secure to me.

Quote:
Also a little while back I was able to get two tubes, one a raytheon and the other a tung-sol. Both have the halo's. Will I have any problems?
Seeing as how they were both probably made by RCA, I doubt it.
Reply
post #3220 of 6115
5/22/09 at 11:41pm
Quote:
Originally Posted by Logistic View Post
<snip>I chose this tube because it was very inexpensive...</snip>
What??? Logistic please don't tell me this is only reason you chose these tubes

It almost sounds like you could've picked any tube as substitute for 19J6, but you didn't in the spirit of the "starving" theme (?)
Reply
post #3221 of 6115
5/23/09 at 2:14am
Quote:
Originally Posted by zkool448 View Post
What??? Logistic please don't tell me this is only reason you chose these tubes

It almost sounds like you could've picked any tube as substitute for 19J6, but you didn't in the spirit of the "starving" theme (?)
Of course I didn't just pick this tube because it was inexpensive. While there are a number of tubes that would work here, there are a lot that wouldn't. When searching for a tube, the cost was only one of the factors I considered, and only after I made a list of tubes that might work in the amp.

It is possible to adapt other tubes to the amp, but my primary motivation was to provide an alternative to those would couldn't afford to pay a premium on the scarce 19J6's. Fortunately TomB is providing some 19J6's at beezar, but the supply is limited. This is such a great project that I wanted people to be able to build it that may otherwise have trouble with acquiring tubes.

Logistic
Reply
post #3222 of 6115
5/23/09 at 2:41am
Quote:
Originally Posted by tomb View Post
I'm one of those swimming against the stream, I guess - I don't like threaded and tapped heat sinks. Actually, the pins should work quite well if you use a copper ground plate as Pete used in his original. Look carefully at his pics and you'll see he drilled the holes for the pins, pushed the heat sink pins through, and soldered washers on the back side to the pins and to the plate. Looks pretty secure to me.

Seeing as how they were both probably made by RCA, I doubt it.
should the heatsinks be grounded or does it really matter?
Reply
post #3223 of 6115
5/23/09 at 3:50pm
Quote:
Originally Posted by MoxMonkey View Post
should the heatsinks be grounded or does it really matter?
Not really sure if the recommended practice is to always ground the heatsinks (safety reasons?), but if they are confirm your MOSFETs are electrically isolated so make sure to use TO-220 mounting kits that have some kind of insulators/shoulder washers.
Reply
post #3224 of 6115
5/23/09 at 10:00pm
Quote:
Originally Posted by zkool448 View Post
Not really sure if the recommended practice is to always ground the heatsinks (safety reasons?), but if they are confirm your MOSFETs are electrically isolated so make sure to use TO-220 mounting kits that have some kind of insulators/shoulder washers.
Yah they are isolated with shoulder washers already but no idea if grounding the heatsinks would be even safer.
Reply
post #3225 of 6115
5/24/09 at 3:00am
Quote:
Originally Posted by MoxMonkey View Post
Yah they are isolated with shoulder washers already but no idea if grounding the heatsinks would be even safer.
The more metal the better when it comes to grounding. I would choose to always ground the heat sinks, but that's me.
Reply
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