Generic Open Chassis Design ... - Page 2 - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

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 10th January 2005, 08:42 PM   #11
diyAudio Member
 
BobEllis's Avatar
 
Join Date: Sep 2003
Location: Upstate NY
I think that you may be confusing output power and dissipated power. Chances are good that your 600 watt pioneer isn't dissipating more than 75 watts on an average basis. KWski is referring to class A amps and heavily biased class AB, ala Pass. Ths type of amp dissipates a lot of heat continuously, while most commercial class AB designs idle at a few tens of watts and since most people aren't going to use more than a few watts RMS, the average load is low.

My A75s have over 600 square inches of heat sink surface area per channel (10"h x 13" d x 1.5" thick, 20 fins)and have no trouble at all maintaining 55C on the heat sinks, idling at 200 W/channel.

If you are talking about a sealed chassis and trying to conduct several hundred watts off each amp and dissipate it on the chassis, your devices will get way too hot to have a useful service life. At each junction there is a thermal resistance, due to the surface not being perfectly flat. Sheet metal doesn't have enough cross section to conduct much heat efficiently. all this results in a tremendous temperature rise from the heat sink to the device junction.

Take a look at the AXE-1 calculator on the AX wiki - it includes a spot where you can add the thermal resistance of your transistor mounting scheme. Let's be really generous and say each interface has a thermal resistance of .1K/W. card-rail-case-sink is three junctions, so put .3 in that spot. Don't forget the .7 onto and .7 off the device insulator. See what that does to your junction temperatures.

The only way that I can see your idea working for class AB amps is to include ventilation and a space for each card to have its own heat sink, something like what outlaw audio does here: http://www.outlawaudio.com/products/770_gallery.html# Sealed up the way it sounds like you're heading, I agree with Kilowatski, useful for class D and low power chip amp applications only.
  Reply With Quote
Old 10th January 2005, 08:43 PM   #12
diyAudio Member
 
BrianDonegan's Avatar
 
Join Date: Dec 2004
Location: TPA HQ
I'm thinking you want a standard bus connection for the card edge. One connector uses a three pin high amp connector for +V,-V,Gnd. I guess this is your buss bar idea, but should it be pluggable? could always machine copper bits that mount on the card, then screw to the bus bar from the rear with copper screws. Copper does corrode, but a tight screw connection will not corrode where is matters, not for a long time anyway.

As for heat sinking, could just add custom heat sink cards adjacent to hot modules. Can have micro fans blowing out the back or even water cooling. Pull the heat from the chassis at module edge. Maybe just make your module 2 or 3 u wide by bolting together.

Another possibility. Modules are inserted from the top. Top of enclosure uses sections - flat metal or heat sink depending on what's in a given slot. or the heat sinks can slide onto the back from the top - picture an aluminum dovetail or something on the back of the module with the devices mounted to it: slide in your mode, screw to buss bars for power, slide on the heat sink (which can couple to the adjacent module spaces to make it larger).

Speaker connections on the bus as well? Comes to a problem with how many speakers you can output to per chassis. Probably best to have speaker outs on the back of the cards.

Just stormin ideas.

-b
  Reply With Quote
Old 10th January 2005, 09:11 PM   #13
diyAudio Member
 
BrianDonegan's Avatar
 
Join Date: Dec 2004
Location: TPA HQ
Forgot to mention liquid cooling. Makes me cringe too, but there are a lot of water-cooled products out there now for computers. Cooling blocks that sit right on the CPU, use a small pump, and pump to a radiator (with or without fans).

Could be the top plate or back plate is a big heat sink with water cooling built in. This keeps it sealed and away from direct contact with the devices.

-b
  Reply With Quote
Old 11th January 2005, 01:23 AM   #14
diyAudio Member
 
kilowattski's Avatar
 
Join Date: Aug 2003
Location: Long Island, New York
TVP,

No way on the heat sink. Do your calculations; 130 degrees F is 54 degrees C. That means your heatsink needs to support a 34 degrees C rise above ambient. If you do the division for the 500 watts that means you need a heatsink that can handle 0.68 degrees/ watt. Do you know how large of a heatsink that would take? You would need a 17" piece of the following:
Attached Images
File Type: gif kl-273z.gif (13.9 KB, 359 views)
__________________
-----------------------------------------------
Kilowattski
  Reply With Quote
Old 11th January 2005, 04:08 AM   #15
TVP is offline TVP  United States
diyAudio Member
 
Join Date: Jan 2005
Location: GA
Hey All .,...

thermal dynamics 101

If you READ my whole thread ... I'm NOT trying to dissapate the heat across ONE LAYER of SHEET METAL ... there are ATLEAST 4 layers of sheet metal there ... and an offer of a heat sink for the top AND the bottom not to mention fans??? and why do you THINK you can't dissipate 375 watts ACROSS a SQUARE FOOT ????? you know .. that's the CROSS SECTION in this application ??? I'm not cutting the sheet metal into strips and trying to tranfer the heat from one edge to another edge .. I'm trying to tranfer the heat from ONE FLAT SIDE OF THE SHEET TO THE OTHER FLAT SIDE ... right ????

ahhh ... no ... I'm NOT confused ... I was on the team of poeple that managed to put a 100 gig optical router hanging out on a telephone pole ... SOMETHING ALL THE EXPERTS IN THE WORLD SAID IS IMPOSSIBLE ...yet they have been hanging there and running for OVER 4 years !!! I often do the impossible .. every day infact ... oh , and BTW .. it's dissapating 118 watts of HEAT in the NV AZ Sun AND you can touch the case !!! look ma no FANS either !!!

so your saying your A75's aren't UL listed ?? i.e. 30 deg C is the MAX allowed case temp from UL ... or are you saying they are 55 d C AT the transistor ??? and get cooler from there ??? they idle burning 200 watts of HEAT or even 200watts or 110 line AC ???? holy **** bat man ... what are they total power rated !!!!

so a car can't dissapte 250k ++++ BTU's (500kw++) through a single layer sheet metal radiator ???? heat dissapation is a matter of square inches of dissapation area ... more sq inches = more heat transfer !!!

unless you guys are talking (behind my back) about putting 3-4kw with 80% eff into this chassis .. I really don't think heat is the issue to be concerned about ...

size is the problem right now .....

power ....

yes the idea is a bus bar here .."just" a simple piece of copper or brass or whatever (gold plated .. I'm thinking of a peice of 1/4" stock with a U shaped clip (on the module) that wraps around 3 sides of it ... running the WHOLE width of the module ... sounds like we need 5 supply rails ... 3 high current , higher voltage .. with 2 small rails for the lower voltages ????

short of 3-4kw I don't think we'll need a water system on this either ... don't cringe .. they use that almost exclusively on ion gas lasers .. Argon , Krypton etc ... with a VERY HIGH VOLTAGE AND CURRENT SUPPLY !!! a heat pipe or freon system would accomplish the same goal without the electrocusion hazard

other ideas I've been tossing around ...

spring load the bottom of the chassis so the modules have some tension on them ... this increases the thermal transfer rate ... simluar idea for the front to rear of the modules .. this puts pressure on the power rails and increases the electrical power transfer rating as well ....

keep 'em comming ........

thanx

TVP
  Reply With Quote
Old 11th January 2005, 04:10 AM   #16
SY is offline SY  United States
diyAudio Moderator
 
SY's Avatar
 
Join Date: Oct 2002
Location: Chicagoland
Blog Entries: 1
TVP, your enthusiasm is commendable. A bit less shouting might help make your point better.
__________________
The more you pay for it, the less inclined you are to doubt it.- George Smiley
  Reply With Quote
Old 11th January 2005, 04:42 AM   #17
lgreen is offline lgreen  United States
diyAudio Member
 
lgreen's Avatar
 
Join Date: May 2003
Location: San Diego, USA
Default getting good

People here think they know everything, a bunch of engineers right?

Shout as much as you want, I love reading this thread! Please post some plans or pics when you get far enough along.
  Reply With Quote
Old 11th January 2005, 11:10 AM   #18
diyAudio Member
 
Join Date: Dec 2004
Location: South of Brisbane Metro, Queensland.
I sorry to be a wet blanket on the whole thing but, there were a number of commercial card bassed amp systems running arround the pro market in the early eighties. Currently a lot of fire evac systems run on card bassed systems.
Some of these systems wer designed for flexibility of installation .... option. ect. others were designed for rapid field replacement of faulty modules.

My impression has been
1. That all the fiddle & modules..... are much more bother and much more expensive than starting from scratch. Afterall all you need is a transformer, bridge, a few big caps, an amp pcb or two & some heat sink... oh add some connectors.
2. all the modularising stuff introduces "trouble" extra connectors all the stuff you don't need and most " audiophiles" are trying to minimise.
3. every now & then these ideas pop up & even get to market & disapear just as fast.


I have seen quite a bit of audio stuff that "looks" like it was designsed by somebody with a background in heavy comm's or instrumentation & thats all very nice & the workmanship in design is wonderfull. But its wasted on audio power amps.

There are heaps of transformers out there, they come with their mounting hardware.
A 35 amp bridge is a simple common part & mounts with one bolt.
Biiiggg caps come with mounting gear, easy.
All the electronics can go on one board & you hang that off your heatsink.
Screwing it into a rack box is a very simple task.

Sorry fellas I think its a waste of time & brain power.
  Reply With Quote
Old 11th January 2005, 11:15 AM   #19
diyAudio Member
 
BobEllis's Avatar
 
Join Date: Sep 2003
Location: Upstate NY
200 watts per channel is a lot of heat. That was my point about class A operation - you're lucky to get 50% efficiency. The Pass Alephs go around 33%, Zens can be 5% efficient. Some of us think that the sound quality of class A operation is worth the heat. My 200 watts (400 W stereo) idle gets me 50 watts out if I limit it to class A, maybe 150 if I go AB, since my rail voltages are higher than prescribed.

No, I don't try for a UL rating on my DIY projects, although I follow practices to make them safe. You CAN touch my cases, but not for long.

Even spreading your heat across the flat surface, you've got a couple of interfaces to cross - grease will help but the larger surfaces make it that much harder to get it flat across the width, even for a professional machine shop.

Your router probably has its power devices mounted directly to the inside surface of the case, and they are massively parrallel - lots of devices carrying the load mean the junction to sink thermal resistances parallel, just like electrical resistors. Even so, I suspect that you are probably pushing the junction temperatures a bit higher than the DIY reliability "limit" of 100 C in the summer sun. Nothing wrong with that, just don't expect it to live 20+ years. Again, I refer you to the AXE-1 spreadsheet to see the effect adding more devices in parallel has on junction temperatures.

Heat dissipation capability is a function of surface area AND mass flow across it and the thermal capability of the medium to which you dissipate the heat. you need more area in space, since there is no convection.

Here's a problem I was part of a team that solved: Dissipate 5KW on VME format cards at 70,000'. Don't forget ram air temperature rise. You need a fan for low speed/altitude ops, but at high altitude, the fan will impede the airflow. You're not the only one who has worked on some difficult problems. People smarter than me did the solving, I can regurgitate the solution, but I'll probably miss some important detail.
  Reply With Quote
Old 11th January 2005, 12:05 PM   #20
diyAudio Member
 
BobEllis's Avatar
 
Join Date: Sep 2003
Location: Upstate NY
lowlevel has an excellent point - in a production system requiring flexibility and (field) upgradability, the card bus makes sense. However, in a one off (or even ten off) project it's just a lot of expense with no real benefit and maybe even a detriment. It's way too easy for us engineer types to get carried away with the "wouldn't it be cool if..."
  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
Open call for suggestions on Open Source DIY Audio Design gfergy Everything Else 1 15th April 2007 07:33 AM
chassis cad design hcbonfim Parts 0 29th November 2006 09:28 PM
Gutted amp on an open chassis... EchoWars Solid State 2 5th April 2005 03:34 AM
Generic Open Chassis Design ... TVP Class D 0 12th January 2005 03:21 PM
Generic "Open" Chassis Design Effort ... TVP Chip Amps 0 11th January 2005 04:08 AM


New To Site? Need Help?

All times are GMT. The time now is 02:47 PM.


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