Having an issue with a MOSFET-MAX build
Apr 11, 2012 at 9:14 PM Post #61 of 72
It's a Radio Shack 2200813 and it was reading "0.F".
 
As a side note after turning RB12 L/R COUNTER CLOCKWISE these pics were able to be taken...
 

 
These were taken approximately 10 minutes ago. The LEDs are still shining, the little heaters in the top of the tubes are still burning.
 
Now two things. 1) I have not begun biasing yet so RB12 L/R is still completely closed
 
                          2) And this may have more meaning. The first few times I powered this up and it ran for a bit before dying, the 2 MOSFET heat sinks in the middle were CRAZY HOT. I could only briefly touch them. Now, and like I said I haven't touched the trimpots, only the LM317 is a little warm. The 4 MOSFETs are cool to the touch.
 
This all sound better? Going on well over 15 minutes now FYI with no blown fuse, bright LEDs and cool MOSFETS...
 
Apr 11, 2012 at 9:25 PM Post #62 of 72
Well, that is a good sign indeed.

I would proceed to biasing the beast and see if you can make it sing.

Do me a favor when you get a chance (low on your care-about scale right now I am sure :D ). Measuring for resistance, touch the two DMM probes together in mid-air and let me know what the display says. Then, take them apart and let me know what the display says.
 
Apr 11, 2012 at 9:30 PM Post #63 of 72
Lol, both tubes biased to about 12.5V right now (they're a little jumpy so I'm trying to keep it low).
 
On resistance setting when I touched the probes together it started at 0 then slowly raised a bit. Then I moved and the numbers jumped all over the place but I didn't see it jump over 6 Ohms and that was only for an instant.
 
Apr 11, 2012 at 9:31 PM Post #64 of 72


Quote:
It's a Radio Shack 2200813 and it was reading "0.F".
 
As a side note after turning RB12 L/R COUNTER CLOCKWISE these pics were able to be taken...
 

 
These were taken approximately 10 minutes ago. The LEDs are still shining, the little heaters in the top of the tubes are still burning.
 
Now two things. 1) I have not begun biasing yet so RB12 L/R is still completely closed
 
                          2) And this may have more meaning. The first few times I powered this up and it ran for a bit before dying, the 2 MOSFET heat sinks in the middle were CRAZY HOT. I could only briefly touch them. Now, and like I said I haven't touched the trimpots, only the LM317 is a little warm. The 4 MOSFETs are cool to the touch.
 
This all sound better? Going on well over 15 minutes now FYI with no blown fuse, bright LEDs and cool MOSFETS...


Yeah, it sounds like mounting those trimmers backwards (a good thing for your casework, btw) may have fooled you into turning the trimmers the wrong way to supposedly put them at -zero-.
 
If they were wide open, the buffer and MOSFETs would've drawn enough current to blow the fuse every time.  Your description of "CRAZY HOT" confirms this.
 
Go slow and methodically for the rest of the way.  Review the Setup Page on the MOSFET-MAX website.  The good thing about MOSFETs is that they're like switches.  When you go to adjusting those RB12 trimmers, you will easily see on your DMM (set to measure mV only!!!) when the MOSFETs switch "ON."  From there, you can gradually increase them to the 90-100ma minimum (measured as 200-220mV on your DMM!!!).  If everything checks out, you get sound in your headphones, etc., and the tubes don't go haywire after a few minutes, then go ahead and adjust the bias upwards to 125ma.
 
Let us know how it goes ...
 
 
 
Apr 11, 2012 at 9:43 PM Post #65 of 72
Lol, both tubes biased to about 12.5V right now (they're a little jumpy so I'm trying to keep it low).
 
On resistance setting when I touched the probes together it started at 0 then slowly raised a bit. Then I moved and the numbers jumped all over the place but I didn't see it jump over 6 Ohms and that was only for an instant.


Quirky little meter you have there. Ah well, at least it appears to be getting the job done. I am going to assume the ".F" indication indicates infinite. That would make sense I suppose. I at one time had a Sperry meter that did something similar. Atlas, it is currently (no pun intended) lost somewhere up buried in the insulation of our attic.

Be aware, the tube bias is going be all over the map til they settle down. It is easy to get obsessed with them being set "perfect". After a few days of use, they should stop kicking and fighting and you can spot check if you like to indulge any latent OCD leanings that most DIY'ers seem to possess. :wink:
 
Apr 11, 2012 at 10:05 PM Post #66 of 72
Lol yeah it's quirky but nowhere near as bad as the goofy little thing that it replaced.
 
One more question and then I bid you all goodnight til next time.
 
- Accessing those test points with those tiny little probes and trying to keep a good connection is a giant PITA. I can see where those test points I see installed on the back of MAX enclosures can be very useful and I'm going to find a place to source them out from ASAP but in the meantime, I would think soldering a wire to the test points on the PCB and then soldering any old metallic object on the end of the wire for the probe to fit in would be a lot more accurate and probably calm down some of my readings. I'm assuming more contact area = more accuracy? Just seems like when I'm trying to touch the probe to the pad, with what little contact area is making connection, if my roommate sneezes my reading jump...
 
Oh yeah so that's the question. Assuming I label everything properly, would there be any reason not to get something suitable at Rat Shack to properly bias this beast until I can source out, order and wait for those test points to ship?
 
Guys once again and I cannot say this enough, thank you so much for all the help, patience and dedication. As soon as my order comes in from all-spec containing a .80mm chisel soldering tip I can start murdering, errrrrrrrr building my grubDAC kit...
 
-I'll keep everyone posted and hopefully within the next week or two have something to share on the MOSFET MAX thread (it's late but I seem to recall seeing one?)...
 
-Chris
 
Apr 11, 2012 at 10:28 PM Post #67 of 72
Well, just be careful with metallic objects hooked to wires hanging off your finally functional amp. I would recommend against it. Unless you want more practice replacing damaged parts. :wink:

You can always wrap a paper clip around your probes and then wrap that in electrical tape so that only a tiny portion of the end is exposed. They tend to fit well into those thru-holes. Certainly cheaper and much less prone to accidental shorting.

You want style "D", although any of them work fine. Style D is the recommended tip jack. Come in all kinds of colors as well.

http://www.mouser.com/catalog/catalogUSD/645/1339.pdf

Look forward to seeing the final product in the main MOSFET-MAX thread.
 
Apr 12, 2012 at 8:32 AM Post #68 of 72
You may have been "burned" by a "fire" that shouldn't happen again and are being unnecessarily gun-shy.  When you are reading voltage, the DMM assumes a super-high resistance in order to not affect the readings.  You would have to try very hard to short something by sticking the probes in there - and the test points don't qualify.  Seriously, the tip jacks at the back are there so that you don't have to open the case back up when you've finished it, not because of a danger of shorting.  They are there for convenience.
 
Like Jake says, I'd almost be more concerned at you having wires dangling off the PCB that could come into contact with each other or something else.  I've never heard of someone shorting a MAX when they're measuring voltage.  Not saying it can't be done, but you have to try real hard.  You would have to lay a single probe's metal tip so that it contacted two points at once.  Unless you're laying the probe flat against the solder joints on the bottom of the PCB - even that's impossible because of the plastic finger ring around the end of the probe (it won't lay flat) and I doubt seriously that anyone is going to fire up a PCB and then turn it upside down.
 
Anyway, it was your meter in the current mode that caused this.
 
Apr 12, 2012 at 10:02 PM Post #69 of 72
I'm hoping you guys haven't had the thought "Damn, I'm wondering what the Nomnomnom guy messed up tonight" lol especially after my last post last night.
 
- Just to ease the least of your thoughts I had no visions of crazy little exposed wires laying around ready to ground themselves out to anything, the only thing that spawned my post last night was a little thing that could tide me over until what I hope is to be my last order from Digikey or Mouser (but they killed me on shipping last time lol). I'll spend tomorrow after work figuring out exactly what I need to complete this project so I can place one final order (to include whatever I may need for the grubDAC).
 
- I haven't posted anything tonight because.... well.... I've been listening to some amazing things out of my headphones (T50RPs and 225Is) for the past couple hours. Is there nothing this amp can't do? The dynamics are absolutely insane, I pushed my poor drivers and my ears to the MAX (
L3000.gif
) and neither failed (I only pushed my ears so far). I've never heard such authority from either of
 
my headphones.
 
Guys, I'll say it again and again if need be. You've helped me through this when I had no clue and I appreciate the hell out of it..... Thank you a million times over....
 
Voltage is reading a very steady 27V
 
Tube bias is right around 13.5V but it's a little jumpy (settling nicely though)
 
MOSFETS are at an awesomely steady 260mV (I'm keeping a close eye on them, the tubes and the voltage)
 
Nothing is "CRAZY HOT", offensive to my olfactory senses or sending fireworks into my vision.
 
I plan on compiling all of my experiences together to write a fairly explanatory new thread at some point and time over this coming weekend. I plan on aiming it at noobs like myself that may have a few questions, maybe some of the basics that they (like myself) overlooked. I have also started some plans on casing this beast up but I really want to make something special so it may take a bit.... Oh yeah, I definitely have to assemble my grubDAC and see how that sounds too. For some reason I don't foresee disappointment in my future lol. My shipment from all-spec with some SMD tools (tweezers and .80MM chisel tip) should be in tomorrow so.... Some more fun tomorrow night....
 
Everyone in this thread that has helped me through this is FREAKING AWESOME. You are experts in your field (hobby?) and are to be commended. I thank you a million times over BUT... I do have one thing to say to all of you.... None of you has welcomed me to Head-Fi with the obligatory "sorry for your wallet"...
tongue_smile.gif

 
Not worried FYI, I've already apologized to the poor old leather guy a million times over hehehehehe....
 
-Chris
 
Oh crap and a pic explaining the non-conductive temporary solution to my biasing needs and I have to say. The probes fit perfectly into these "12 position - Female - Interlocking Connector" (#2740242) from Radio Shack and I labeled the inserts with permanent marker covered by clear tape. Frees the hands up and clears the test areas to turn the trimpots to your hearts' desire!
 

 
Yes, that's electrical tape around the power socket (I try to learn from my electrical mistakes lol). The T50RPs sound absolutely awesome and my SR225Is have never sounded so.... Grrrrrrrr a word considered to be a swear would be perfect here but "engaging (?) will have to suffice.
 
From what I hear I really can't wait to assemble my grubDAC kit. If it's better than the uDac2 then hmmmmmmm, maybe that case work will have to wait a little longer lol......
 
Apr 13, 2012 at 1:54 AM Post #70 of 72
Glad to hear you're having fun with it!  You've got two of the best headphones that a MOSFET-MAX can drive: Grados and orthos.  Looks like your T50rp's are highly modded, that usually means more current is needed than normal.  The MOSFET-MAX can provide that in droves.
 
I think you will like the GrubDAC much better than the uDAC, but let us know (I don't know for sure).  A flux pen and a smaller tip for your soldering iron will do wonders in assembling the GrubDAC.  Don't panic over it if it doesn't work right away.  The GrubDAC thread can give you a lot of help and cobaltmute is excellent at troubleshooting.
wink.gif

 
Apr 13, 2012 at 3:48 PM Post #71 of 72
Nice work Chris.

Be aware that when you finish and install in the final case, the tubes and buffers will need to be rechecked due to the change in thermal characteristic. Just a heads-up.

I bet you never in a million years expected "sorry for your wallet" to apply to shipping costs. Get used to it. No matter how much preparation, no matter how much checking and rechecking, there is ALWAYS something you need. You might want to seek out and see if there are any small electronics shops in your area. If your really lucky, you might be within driving distance of a Fry's. Parts quality there is a mix, but the convenience can't be beat. Regardless, a local source can take the sting out of reorders. You can also ask on this board if someone has an extra part or two to meet an immediate need. Most can be shipped for the cost of a stamp.

That Molex connector is a clever idea. In fact, a six pin version of that part might make a nice, compact permenent solution. Just have to figure how to mount it appropriately.

Enjoy your amp. I know you will. :)
 
Apr 14, 2012 at 8:02 PM Post #72 of 72
Hey fellow Head-Fiers,
 
     A little progress report:
 
- GrubDAC assembled! But not functional
frown.gif
 . I'll be posting details and pics sometime tomorrow over in the GrubDAC thread (I really just feel like listening to music and not messing around with anything tonight). For anyone reading that has never done SMD work I'll suggest the same thing that I was told in an advanced marksmanship class. LAY OFF THE CAFFEINE! Lol, these parts are impossibly tiny and we as humans freakin shake so take whatever precautions you can before attempting to hold these miniscule parts on the board and, heed TomBs advise, flux and a small (I had already ordered the 0.8MM chisel tip) are the only way to attack this project. I didn't have a flux pen but I found that a little jar of rosin flux paste and a somewhat stiff, small paint brush as an applicator worked wonders (a hobby paint brush not one that you use to paint the trim on your door).
 
- MOSFET MAX still kicking strong! I haven't let it "burn-in" per-se only for fear of electrical fire in Army barracks but everytime I listen to it I have my trusty DMM and small flat blade nearby and take a few minutes to bias everything from V+ to the MOSFETS. This thing is still absolutely awesome, I go back and forth from my SR225Is to my T50RPs depending on what I'm listening to and what I want to hear and DAMN!!! Awesome, absolutely awesome...
 
- Tom, I did a butt ton of research on what amp would provide the proper current/voltage swing(?) for my needs and that combined with my incessant need to make stuff led me to the MHM and despite my noob-ish issues I have no regrets, this thing rocks! I did purchase the smaller tip (stock Hakko tip is a monster!) and had plenty of flux paste at the ready and still managed to produce a non-working product but like I said, I'll enjoy the fruits of my labor tonight and address the others tomorrow. Thanks again both for your service (Beezar) and support.
 
- Jake, I do recall reading something about a "temperature coefficient(?)" of the MAX cased vs. not cased but thank you for reminding me and thank you very much for the compliment, I couldn't have done it without you guys. Lol that damned shipping cost... I'm pretty sure it was Mouser that, when finalizing my order, stated that the shipment cost would be calculated and added to the cost after completion and, for a normal ground delivery, ended up killing me for like $17!! I have a slightly bad taste in my mouth from that experience and I'm hoping I can find everything I need to complete this project (my last? HA!) from Digikey. I'm located in Fort Drum, NY so the best I have is Radio Shack, no Fry's or anything other than BestBuy and farm equipment up here... I was looking at the 6 pin molex connector BUT my (and I think pretty much everyone else') probes have those safety guards on them preventing the 6 pin model from being useful. The 12 pin that I bought was perfect. Top 4 = V+, TB2L, TB2R, GND. Middle 4 = not used, Bottom 4 = TA2L, empty, empty, TA2R. They are all spaced perfectly so that the test probe guards never collide (very close though!) during biasing. I may research single pin versions becasue if such an animal does exist, it may fit my plans for casing (which are al over the place!).
 
-- BTW I have the specified Hammond case in silver but my brain is throwing some pretty crazy ideas involving it, wood and Lexan for the final product and as much as I'd like to say "Look for pics in the MOSFET MAX thread next week" I know myself all too well and between my ever changing mind, access to the on-post wood shop and free time, it may be a bit before this beast makes it over there but I'm going to try like hell to make it as picture worthy as it is listen worthy....
 
Thanks once again everyone,
 
-Chris
 
 

Users who are viewing this thread

Back
Top