My parts arrived today and I went through George's list.
Apparently all that was wrong was a blown FRED. The mosfets appear fine. The FREDs are exactly the part specified on the web site - they appear to be IXYS parts. I really hope they don't blow again.
I'm making a few modifications to the amp now that it is apart. I decided to add a small 12v fan to blow across the heat sinks and the board. I'm going to isolate it with some rubber grommets, and power it with a small wall wart that I've taken apart.
I also found a small heat sink that I'm going to mount on the back of the existing D1/U1 sink - between the larger heat sink and the fan, I think I should have a much cooler chassis.
I should have everything put back together tonight...
Apparently all that was wrong was a blown FRED. The mosfets appear fine. The FREDs are exactly the part specified on the web site - they appear to be IXYS parts. I really hope they don't blow again.
I'm making a few modifications to the amp now that it is apart. I decided to add a small 12v fan to blow across the heat sinks and the board. I'm going to isolate it with some rubber grommets, and power it with a small wall wart that I've taken apart.
I also found a small heat sink that I'm going to mount on the back of the existing D1/U1 sink - between the larger heat sink and the fan, I think I should have a much cooler chassis.
I should have everything put back together tonight...
The amp is up and running again!
And I have a much better thermal solution - although the fan I'm using can be heard with the cover open, once I get the cover closed up I'm hoping that it will be fairly quiet.
With the new heat sink and fan the D1/U1 combo are much happier. Roughly 1/2 the temp of before - 40 to 50C.
pic below
And I have a much better thermal solution - although the fan I'm using can be heard with the cover open, once I get the cover closed up I'm hoping that it will be fairly quiet.
With the new heat sink and fan the D1/U1 combo are much happier. Roughly 1/2 the temp of before - 40 to 50C.
pic below
Attachments
Ok, I noticed this before, but one of the channels comes on first by a few seconds... I assume that this is the difference between the two input tubes. But it's pretty abrupt, not like a heater gradually coming to temp...
Oh - and THANKS for all the help George, Russ & PaulyT... it's great to have support...
Oh - and THANKS for all the help George, Russ & PaulyT... it's great to have support...
You're welcome, though I don't think my "help" was as useful as Russ's and George's... There will be time to return the favor, since I'll hopefully be powering up mine pretty soon, and I'm sure I'll have mor questions. I'm having a hard time finding an appropriately sized heatsink, too, though I think I've got one ordered now that will work.
Yeah, I was just kidding about the fan. 
I got a heatsink on the large side so that hopefully I won't need a fan, especially since my enclosure is going to be relatively open; just that the one I initially ordered was ridiculously large, as was the old cpu heatsink I had lying around. I was afraid it'd put too much mechanical strain on the small soldered-on heatsink and the ICs and PCB.
I got a heatsink on the large side so that hopefully I won't need a fan, especially since my enclosure is going to be relatively open; just that the one I initially ordered was ridiculously large, as was the old cpu heatsink I had lying around. I was afraid it'd put too much mechanical strain on the small soldered-on heatsink and the ICs and PCB.
PaulyT said:Yeah, I was just kidding about the fan.
I was initially going to use a Pentium heat sink, but mine wasn't going to be supported by the devices themselves. It sat on the bottom of the chassis, and contacted the existing heat sink via spacers. The problems I had with that arrangement was that I still needed to cut a channel in the heat sink for the PCB (it was still too far away from the heat sink) and I had no way to mount it - drilling a hole in a finned heat sink was just going to break a few bits, and it would still need to be a two step drill, one for the screw body (from the bottom) and one to capture the head of the screw (drilled from the top). Otherwise I'd need a very long screw and washer to make it all the way out to the top of the fins. The best way to do this is to use a thermal tape or epoxy - but it's generally expensive from what I could find.
I settled on the second heat sink to bolt onto the first even though it was smaller because I really do need to move some air through the box. With the top on the whole chassis gets very warm! Russ's solution of using a finned chassis and conducting the heat to the sides is exactly the way that I would do it - if I had the tools and the time/money to implement it.
I know that I've made things even worse by using my UOS (Used Old Stock) 5U4 rectifier. The Edcor power xfmr gets very toasty after a few hours and that is just going to put more heat into the top plate.
I'm sure your convection cooled box will be just fine. Although the look of that giant top mesh cage might be a little odd. And btw, I bought one too for this chassis, but I like the way it looks without it, and since the cage is steel - it doesn't come close to matching the aluminum. I doubt that I could paint it any color that would make it look good...
Yeah, I did manage to mount the old pentium one, just as you said - small holes in the bottom plate, larger holes through the fins on top. It was possible with a drill press, but it was tricky. It looks more than a bit ridiculous, though, maybe I'll post a pic in my thread if I get a chance before the new one comes, just for grins. You all will laugh.
Evenharmonics said:
My TSE chassis is relatively compact and by having an extra heat sink for U1, it didn't need a fan.
Why add more noise.
Well I've designed a lot of electronics - workstations, routers, RAID cards etc... and they all needed fans & airflow. With some of the temps that I see on the silicon in this design it just seems normal to me to want to put a fan in. The challenge is to make it quiet!
I actually am considering a peltier module, but I've never used one before - but that would only help on the one heat sink and not the mosfets...
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=102-1665-ND
Try a small squirrel cage fan, like a PCI slot fan. At full speed they are very quiet...much quieter then any 40mm fan. At half speed they are nearly silent. Just be sure to get a ball bearing one. With a little extra lubrication I can get many years of service from a cheap ball bearing slot fan.
oldmanStrat said:
Yeah, if the power dissipation is relatively small and you are using an aluminum chassis, that works well.
Those CPU heat sinks are pathetic without a fan. They are built for operation with airflow. And they do that quite well. However, to me, audio and fans don't mix. I'd rather design the circuit to dissipate less power -- or use a bigger heat sink. But hey.... Maybe that's just me.
~ Tom
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.