Millett "Starving Student" hybrid amp
Jun 22, 2009 at 1:07 PM Post #3,466 of 7,277
Well, it may look a lot worse due to the photo. See if you can get us one without as much glare and maybe we'll be able to tell what's going on.

As for attaching the ground to the plate - solder your wires to a flat washer, then drill a hole in the plate and use a screw and nut to clamp the washer to the plate. Scrape the metal all around the hole where the washer contacts to ensure that some coating/oxidation doesn't insulate the connection to the washer. Use a couple/several washers if necessary and put them on both sides of the plate.
 
Jun 23, 2009 at 4:16 AM Post #3,467 of 7,277
Well, I think I may be putting this project on hold. At this point my materials have been soldered, re-soldered, and are building up a fluxy residue. I think perhaps the best thing to do is to order the kit. I have a feeling that it will turn out nicer and will be a smoother experience. In the meantime I'll build the bantamDAC from the kit, build up on my skills, and have something to occupy me until the amp kit comes. Thank you, especially to TomB for all your help. I will be back again before the summer's end. Either for more help or with triumphant pictures and happy ears.
 
Jun 23, 2009 at 7:10 AM Post #3,468 of 7,277
I have a question.
When a capacitor fails due to excessive heat, does it block all current, or does it usually short circuit?

I'm letting my brother use the one I have right now, but if theres even a small chance that the coupling caps will short circuit, I'd rather just make another one. I made the mistake of thermally-adhering the heatsinks to an small, all-aluminum case, and the case turns into a easy-bake oven.

Its been working fine up until now (and sounds absolutely wonderful might I add), but better safe than sorry =/
 
Jun 23, 2009 at 12:05 PM Post #3,469 of 7,277
Quote:

Originally Posted by JamesL /img/forum/go_quote.gif
I have a question.
When a capacitor fails due to excessive heat, does it block all current, or does it usually short circuit?

I'm letting my brother use the one I have right now, but if theres even a small chance that the coupling caps will short circuit, I'd rather just make another one. I made the mistake of thermally-adhering the heatsinks to an small, all-aluminum case, and the case turns into a easy-bake oven.

Its been working fine up until now (and sounds absolutely wonderful might I add), but better safe than sorry =/



Guess I've been lucky to only blow one capacitor so far in my short DIY career.
wink.gif
When it blew, it shorted as best as I can remember, but I think it can go either way. Generally speaking, the electrolyte starts oozing out everywhere and since that's electrically conductive, it causes the short.

When I blew mine, it was over-voltage. I don't really know what they do when they're exposed to too much heat, except that the performance can no longer be guaranteed. Just a guess, but perhaps they slow bake to a horrible performance level.
wink.gif
 
Jun 24, 2009 at 9:30 AM Post #3,470 of 7,277
Hello guys ^^

Now that my order is ok, I would like to get some informations about the stuff you use to make your DIY amp... Particularly, what do you use to solder ? Is there a suggested power in watts for the "solder tool" (I don't know how to translate it from French ^^ In French it is "Fer à souder")... Also is there a suggested "tin" (Is it well translated ??)

Thanks a lot in advance !
k701smile.gif
 
Jun 24, 2009 at 9:41 AM Post #3,471 of 7,277
Tritonpad:
My guess is that based on this question, you have no DIY experience. In that case, I would suggest an adjustable iron and rosin-core solder. The rosin helps the solder to flow easier, which means you will spend less time with your soldering iron on the joint which is a good idea.

If you don't want an adjustable soldering iron, I suggest 30 watts at the most to avoid doing damage to components. You may also choose to use silver bearing solder if you want.

Hope this helps.
 
Jun 24, 2009 at 9:45 AM Post #3,472 of 7,277
Thank you for that ^^

Yes it helps ^^ Even if I am not completely new in soldering... I just want to be sure not to "kill" something... We are dealing with nice sounds around here and I know the soldering is important...
 
Jun 24, 2009 at 12:27 PM Post #3,473 of 7,277
Quote:

Originally Posted by Tritonpad /img/forum/go_quote.gif
Thank you for that ^^

Yes it helps ^^ Even if I am not completely new in soldering... I just want to be sure not to "kill" something... We are dealing with nice sounds around here and I know the soldering is important...



I am totally sold on Hakko soldering irons. The Hakko 936 adjustable is a particular favorite around here. It can be purchased for less than $100 at most places. Often, you can find a deal for one that includes several different tips for that price as well. I've seen them as low as $50 at Fry's, where I bought three of them.
wink.gif


If you don't want to spend that much right now, their Hakko DASH irons are some of the best basic soldering irons I've seen. The 15W (#N452JN-V12) is probably all you'll ever need. I got the 20W (#N453JN-V12) and it's fantastic. The extra power is a little hot for PCB's, but it comes in handy if you're working point-to-point or with perfboards and cases. I think you can get one of these for $30 or less. They are well worth the slight extra expense over the cheap pencil-type soldering iron. The tips are quite special on these irons and seemingly last forever with proper care.

For solder, get some "Eutectic" - a 63/37 mix of tin/lead with a rosin core. That's the proper ratio to ensure even melting between solid/liquid. Just the use of this solder alone will make your solder joints look professional.
wink.gif
Something in the 0.30-0.025 in. dia. range is probably best. Thicker just gets in the way of everything and smaller diameters break too often.
 
Jun 24, 2009 at 1:13 PM Post #3,474 of 7,277
Once again Tomb, I must say THANK YOU ^^

Actually I have to say that if you were not around here, I would have passed away without trying building SSMH...
Now that the SSMH kit is ordered, I can hardly wait !
k701smile.gif
 
Jun 24, 2009 at 1:43 PM Post #3,475 of 7,277
Quote:

Originally Posted by Tritonpad /img/forum/go_quote.gif
Once again Tomb, I must say THANK YOU ^^

Actually I have to say that if you were not around here, I would have passed away without trying building SSMH...
Now that the SSMH kit is ordered, I can hardly wait !
k701smile.gif



That's very kind of you to say that! Hopefully, you won't change your opinion once you start building one.
wink.gif
wink.gif
wink.gif
 
Jun 24, 2009 at 3:04 PM Post #3,476 of 7,277
It has been a very long time since I have fired up my soldering iron and made anything. If memory serves it was an early 80's Hafler. My knowlege of power supplies is therefore dated. I remember raiding surplus shops and Ham Fests for very large iron - transformers/chokes and enormous blue caps that could power the oven for a half hour after being unplugged.

Long story short, I think I have goofed up by purchasing a switching power supply in anticipation of building my SSMH. I bought first and am asking questions later. I am learning that these are very sophisticated little buggers and appear not to take kindly to output "filtering" or further regulation. I got a new MEANWELL SR-25-48 48VDC trolling on Ebay and while it is impressively stable in terms of voltage, the max ripple is 200mVp-p.

I am new to these forums and my searches have been like going down rabbit holes. I have been looking for filtering strategies and loading strategies that would optimize a SMPS for this SSMH amp. I think I have painted myself in a corner having only 48VDC coming out of this power supply.

It may end up a cute paperweight but if someone could direct me to information or strategies to clean this DC up I would be much obliged.

Bruce
 
Jun 24, 2009 at 4:03 PM Post #3,477 of 7,277
Quote:

Originally Posted by brucegseidner /img/forum/go_quote.gif
It has been a very long time since I have fired up my soldering iron and made anything. If memory serves it was an early 80's Hafler. My knowlege of power supplies is therefore dated. I remember raiding surplus shops and Ham Fests for very large iron - transformers/chokes and enormous blue caps that could power the oven for a half hour after being unplugged.

Long story short, I think I have goofed up by purchasing a switching power supply in anticipation of building my SSMH. I bought first and am asking questions later. I am learning that these are very sophisticated little buggers and appear not to take kindly to output "filtering" or further regulation. I got a new MEANWELL SR-25-48 48VDC trolling on Ebay and while it is impressively stable in terms of voltage, the max ripple is 200mVp-p.

I am new to these forums and my searches have been like going down rabbit holes. I have been looking for filtering strategies and loading strategies that would optimize a SMPS for this SSMH amp. I think I have painted myself in a corner having only 48VDC coming out of this power supply.

It may end up a cute paperweight but if someone could direct me to information or strategies to clean this DC up I would be much obliged.

Bruce



Pete Millett strayed from conventional DIY-headphone-amplifier wisdom with his use of a switching power supply for the Starving Student. It's worked quite well, but there is little data of successful use of switching power supplies for headphone amps beyond that (AFAIK).

The Cisco PSA18U is the specified power supply for the Starving Student. It's spec'd for 48VDC at 0.38A and performs quite well in the Starving Student. As a VoIP telephone power supply, it's also ubiquitous on ebay and other places, and very inexpensive - sometimes even turning up in surplus and used sales at some outlets.

Short of that recommendation, you are probably best to pursue a linear-regulated power supply for headphone amps. The simplest of these start with an LM317 (TO-220 style) voltage regulator with a bank of electrolytics for filtering and the parts recommendation as provided in National Semi's datasheet for the LM317. Tangent's TREAD (Tangentsoft) is probably the simplest PCB version of such a circuit and can be powered with a simple AC walwart. You can easily achieve ripple values of less than 1mV with this (~55uV?). Other more sophisticated designs include Tangent's new Young-Jung Power Supply and AMB's Sigma 11 and Sigma 22 and the ripple is in single-digit microvolts with those. Cetoole and I achieved 45uV in the Millett MiniMAX's onboard power supply as well - using the LM317 circuit as a basis.

However, we often caution everyone in this thread to think twice before expanding on the Starving Student's design. After all, the low cost is a primary design feature. The PCB version increased costs with the addition of the PCB and the custom case from Hammond, but those are mostly structural changes for convenience in building. Dsavitsk applied a couple of new additions to the actual circuit on the SSMH PCB, but these amounted to a couple of pennies's worth of resistors and a couple of dollars for a couple of cathode bypass caps and output bypass film caps. The changes are well worth incorporating and definitely make the amp sound better, but are trivial in cost and added building effort.

Note that we still recommend the same power supply for the PCB version, however - the Cisco PSA18U.
wink.gif
 
Jun 25, 2009 at 12:29 PM Post #3,479 of 7,277
Quote:

Originally Posted by yebisu /img/forum/go_quote.gif
I've spent a lot of time going over my connections and it looks ok, but I'm not sure why my MHSS still isn't working. I've desoldered and resoldered it to clean it up a little (it's still messy =\), but no luck. Can anyone see what I've done wrong? Thanks!

Flickr: andrewphan86's Photostream



Well, I spent a good while looking at this and you may have the tubes hooked up correctly, but it takes a while trying to trace the wiring in your photos - having the wire all in one color doesn't help, unfortunately.
wink.gif


I can't find R13, but I'm not sure that means anything other than I can't trace the circuit. One thing - aren't the studs only connected to one tab on each of the terminal strips? If so, then it doesn't look like your star ground is connected to ground. All of your leads go to one tab on the terminal strip, but the other tab is connected to the stud. There doesn't appear to be anything connecting the two tabs, so that would mean the star ground is not grounded?
 
Jun 26, 2009 at 2:10 AM Post #3,480 of 7,277
I accidentally used the wrong stud for the star ground, so I connected that star ground tab to the power jack ground. I also isolated the other tab (with the stud) from the case. R13 is hidden under some heat shrink =\ Thanks for you help! I'll keep going through it to see what I did wrong
 

Users who are viewing this thread

Back
Top