Go Back   Home > Forums > Amplifiers > Class D

Class D Switching Power Amplifiers and Power D/A conversion

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 23rd March 2010, 09:27 AM   #1211
v-bro is offline v-bro  Netherlands
diyAudio Member
 
v-bro's Avatar
 
Join Date: May 2006
Quote:
Originally Posted by Zero One View Post

The test board now has 80 hrs up so I won't add the coils or make other changes till 100 hrs is up which should be Thursday. Even then only one thing at a time. That will include Low ESR caps at the chips (as close as possible) optimisation for 26V (roughly what it runs at of the batteries), possible changes to the 5V line and power routing.
41hz AMP4 has the caps miles closer to the chip by default. Note that when placing the caps floating above the board the pin length will diminish the ESR performance.

Quote:
Originally Posted by Zero One View Post
Hey V Bro, I agree about the scope, I don't own one but a good friend who is a retired audio tech does and I am sure he would love to spend some time doing serious testing with me.

As always all input is greatly appreciated.
Okay, well I bet the guys here would love you to post here about your findings!
__________________
Max. cone displacement can be several foot on any speaker!Too bad it can be done only once......
  Reply With Quote
Old 23rd March 2010, 09:47 AM   #1212
diyAudio Member
 
Join Date: Apr 2006
Location: Goulburn NSW
V-Bro Happy to report back on the tests when done.

I was thinking of mounting the caps under the board, where the chip is, I need to check the options, I will lay the caps flat and rotate the caps to get the positive terminal as close as possible, mounting right at the chips I think would be a problem due to heatsink location I think.

Thanks for the tip.
  Reply With Quote
Old 23rd March 2010, 02:34 PM   #1213
wwenze is offline wwenze  Singapore
diyAudio Member
 
Join Date: Mar 2008
Quote:
Originally Posted by Zero One View Post
C) lastly find a way of adding extra pressure to the heat sink without causing any problems with board distortion etc
Washers - put them on the underside of the board and poke the push-pin through them

Quote:
Originally Posted by soldersmoker View Post
What's the approximate pitch of this "buzzing noise" you hear? 60/120 Hz? Or higher? I don't hear any such thing and I have 96dB/1W/1m speakers. Do you have the volume cranked up high when you hear the noise?

I'd like to replace my fan on principle, just to remove a moving part that will eventually need replacing anyway. But I never hear any buzzing.
For me the MSI fan buzz is there with no source connected and lowest gain, the frequency follows the RPM of the fan and I suspect it to be the same numerical value.

Won't get to do anymore test on that now that it's gone forever. The entire heatsink which I left on my table disappeared totally, spooky.

Last edited by wwenze; 23rd March 2010 at 02:42 PM.
  Reply With Quote
Old 23rd March 2010, 02:49 PM   #1214
sendler is offline sendler  United States
diyAudio Member
 
Join Date: Aug 2004
Location: ny
Default Not getting it

Quote:
Originally Posted by Zero One View Post
Hi Sendler My fan is the later ones not the blue one, but it is running off a totally separate battery, so PS doesn't some into it. As far as the noise of the fan itself, it is very quiet, what is able to be heard is a sound that is quiet mechanical in nature though the speaker.

I must emphasis this is very very low level stuff, but I am totally anal retentive when is comes to getting the cleanest sound possible, other wise I would never have built a totally off the grid system (including analogue and digital front ends). I know most folk will never hear a lot of this stuff but my philosophy has been totally eliminate all noise and then optimise from that point on.

There is some low level circuit noise and that will certainly be dealt with but the fan noise is a little louder than that but it is quite possible those listening to amps fed via SMPS will not hear it as it would likely be buried beneath the increased noise of the power supply

I feel that the microphonics (if we can call it that) is indeed related to the heat sink, which is spring loaded to the chips, there is in fact very little effort required to wiggle the sink against the springs. I suspect the heat sink chatters against the top of the chip in sympathy with any external vibration and somehow this translates through the system, by means unknown. Pressing down on the heat sink kills the noise totally.

My intended solution is to A) remove the fan from the top of the heat sink and mount remotely above the chip. B) place small slivers of rubber or balsa between select fins of the heat sink and C) lastly find a way of adding extra pressure to the heat sink without causing any problems with board distortion etc. I will also ensure that all connections, coils and caps are very firmly mounted to the board.

Once this is dealt with residual circuit noise is next on the agenda, to that end all coils are in copper pipe tubes sealed at one end, and there will be double sided copper clad boards on either side of the amp board grounded to a single point. I am also looking at the best place to take the ground off the board, I doubt the current location is ideal, I know from Dacs and Phono pre amps that little things like that can have a reasonable bearing on circuit noise and overall sound.

The test board now has 80 hrs up so I won't add the coils or make other changes till 100 hrs is up which should be Thursday. Even then only one thing at a time. That will include Low ESR caps at the chips (as close as possible) optimisation for 26V (roughly what it runs at of the batteries), possible changes to the 5V line and power routing.

If it all works out then I will build up 6 boards for my 100 db efficient OB system.

Hey V Bro, I agree about the scope, I don't own one but a good friend who is a retired audio tech does and I am sure he would love to spend some time doing serious testing with me.

As always all input is greatly appreciated.
I am definitely not getting any microphonic or piezo electric noise from banging on the blue heatsink with my ear touching the speaker drivers. I'll try one of my third version finned/ fan amps when I get a chance to see if they are any different. Do you have another board? Maybe there is something wrong with yours.
__________________
Scott
  Reply With Quote
Old 23rd March 2010, 02:54 PM   #1215
sendler is offline sendler  United States
diyAudio Member
 
Join Date: Aug 2004
Location: ny
Default Passive cooling

Quote:
Originally Posted by SoldierAnt76 View Post
I can confirm that the heatsink fan is audible through the speakers. When I stop the fan with my finger, the buzzing noise goes away. When I release my finger, the pitch of the buzzing rises as the fan spins back up. So the two are linked.

It isn't a big deal, but it adds to the background noise you hear when the amp isn't receiving a signal. Maybe replacing the fan with a bigger heatsink is the solution?
Your passive volume control and internal wiring is picking up the noise. The easiest and cheapest thing for us all to do is change the heatsinks to a slightly larger passive such as the $6 Zalman.
__________________
Scott
  Reply With Quote
Old 23rd March 2010, 09:57 PM   #1216
teamacc is offline teamacc  Netherlands
diyAudio Member
 
Join Date: Jan 2010
I ran my fan off an 7805 regulator, and most of the noise went away. I think going passive is better though.
  Reply With Quote
Old 23rd March 2010, 11:42 PM   #1217
diyAudio Member
 
Join Date: Mar 2005
Location: Toronto
[QUOTE=



For me the MSI fan buzz is there with no source connected and lowest gain, the frequency follows the RPM of the fan and I suspect it to be the same numerical value.

Won't get to do anymore test on that now that it's gone forever. The entire heatsink which I left on my table disappeared totally, spooky.[/QUOTE]

I got the same result when I first fired up my Sure a couple of weeks ago. The gain is also set on low for the recommended pre-amp setting. I tried it with the fan removed which cured the problem but the heat was getting pretty intense. My next move was to remove the existing wires as close to the fan body as possible and replace them with shielded teflon insulated and wrapped wire I found in a surplus store. While the fan was off, I used aluminum foil tape to shield any part of the thing that wasn't moving. The wires were reconnected to the 5 volt and negative terminal blocks farther from the chip than the original connection and the buzz is pretty much gone. There is a very slight hiss in the left speaker which is connected close to the fan hook-up. With my ear to the right speaker, the hiss is barely audible. With the fan running, the heatsink is very cool to the touch, no meltdown imminent! I was also getting some audible distortion in dynamic bass drum on Kodo's Mondo Head SACD (which is about as dynamic as it gets). I was using a 24 volt, 600ma wall-wart until today when I received a Meanwell clone, 27 volt 13 amp smps from Modders on eBay. In short, distortion is gone and the amp sounds great. My speakers have a double voice coil base driver and need a bit of current, so I suspect that the 600ma was the problem. The amp is driving a pair of Energy Veritas speakers(87db) and the source is a Bryston BP25 pre. It is a great sounding combo, and the Sure is at least equal in hi-frequency and mid range detail to my Bryston 3B-ST ( around 120 watts) but the Bryston will obviously play louder when required and have better control of the bass. The dynamics of the Sure are amazing for an amp of that size and I think the overall sound quality to be far superior to my modded Trends or Sonic Impacts (I have 3 of those). Thanks to all of the leg work from forum members, I am replacing some of the usual suspects on the Sure board. The input caps are Black Gate 4.7 uf N's( contrary to conventional wisdom of no electrolytics in the signal path) because I have a few kicking around, they are small, and I like the sound with Tripath amps. The tank caps are two Pany FC 2200 uF 35v, overkill probably, but again they were what I had. I have some coils coming from Argenhelder because they were mentioned here a few times and they are easy to order on fleabay, quite scientific. The case is a work in progress but I am slowly getting there though I kind of like the bare board look, with wires and caps hanging out.
Attached Images
File Type: jpg IMG_0385.JPG (247.3 KB, 503 views)
File Type: jpg IMG_0387.JPG (237.1 KB, 489 views)
File Type: jpg IMG_0392.JPG (162.1 KB, 475 views)
  Reply With Quote
Old 24th March 2010, 01:04 PM   #1218
diyAudio Member
 
MikeHunt79's Avatar
 
Join Date: Aug 2003
Location: Bristol, UK
Quote:
Originally Posted by v-bro View Post
Looks like it is running on a very high frequency, there's just 1000uF 35V inside now.

Looks like you can drill a few extra holes there in that PCB to swap it for a bunch of smaller caps...

EDIT: I see a 100uF 50V cap on the output as well....
Hi, I eventually got around to finding the switching freqency, with a little trial and error:
Click the image to open in full size.
(apologies for the poor pic)
I measured the voltage on one of the legs of the biggest IC. I think it's either 66KHz or 132KHz, which I think is higher than the Meanwell SMPS, and it seems to sound a little nicer with this SMPS, but there is very little in it.

I'm not really familiar with SMPS's in the same way that I'm familiar with linear PSU's, but am I right in thinking adding too much capacitance would be a bad idea? Are there any calculations or formulas I can use to work out a good figure to aim for?

Right now my plans are to leave the SMPS alone, and have the Sure board with 3000uF of capacitance, so around 4000uF of capacitance in total... I guess thing should be more than enough assuming the switching frequency is 60KHz+?
  Reply With Quote
Old 24th March 2010, 01:34 PM   #1219
diyAudio Member
 
Join Date: Apr 2006
Location: Reykjavík
(Anecdotal: I'm running a pair of 41Hz Amp11's at 20V without any heatsink at all. They have been running like that for at least half a year. No trouble at all. ... I can't imagine why the Sure needs a fan on the heatsink.

By which I mean: The fan is probably not needed at all. Except in a very cramped and closed enclosure.
  Reply With Quote
Old 24th March 2010, 02:26 PM   #1220
sendler is offline sendler  United States
diyAudio Member
 
Join Date: Aug 2004
Location: ny
Default Failures

Quote:
Originally Posted by kristleifur View Post
(Anecdotal: I'm running a pair of 41Hz Amp11's at 20V without any heatsink at all. They have been running like that for at least half a year. No trouble at all. ... I can't imagine why the Sure needs a fan on the heatsink.

By which I mean: The fan is probably not needed at all. Except in a very cramped and closed enclosure.
That's weird because the 4X100 thread reports numerous failures due to the warped piggyback board not allowing a close seating of the heatsink to the outside two chips. And, Sure went to the fan after first selling the 2X100 with a small fanless heatsink and having some failures when used with voltages beyond 24v. I am hoping to run the newest version heatsink without the fan because they do pollute the critical 5v supply unless you install another dedicated supply. I have some amps in the box but haven't tried it. I will measure the temp and report back. I have been wondering if Sure's biggest problems were caused by the thick rubber adhesive they used to use to glue the heatsinks on with which they have gone away from on the newest heatsink.
__________________
Scott
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
FS: Sure Electronics Tripath Board tc2000+tp2050 wimdehaan Swap Meet 9 24th July 2009 11:41 AM


New To Site? Need Help?

All times are GMT. The time now is 06:18 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2