Zero Feedback Impedance Amplifiers - Page 102 - 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 16th September 2006, 05:04 AM   #1011
Sheldon is offline Sheldon  United States
diyAudio Member
 
Join Date: Dec 2003
Location: San Diego
Thanks suiraMB,

No, I don't need much. The problem is that no quantity less than $180 worth is offered.

I have a good multimeter with a thermocouple, so temp. monitoring should be no problem. On the spec sheet, lead soldering for 10 seconds at 260 degrees C is stated, so I would guess that short exposure to the junction of somewhat higher than 150 degrees would not be problematic. In any event, cooking a couple of transistors is a whole lot more cost effective than a couple hundred bucks worth of solder.

I would guess also that the thermal conductivity of the solder, assuming a thin bond, has to be better than thermal grease, or thermal grease and mica. My plan was to solder to a copper plate about 3cm square, then mount that on a thick aluminum sink. I'm looking at 10 watts maximum per device.

Sheldon
  Reply With Quote
Old 16th September 2006, 05:19 AM   #1012
suiraMB is offline suiraMB  Norway
diyAudio Member
 
Join Date: Aug 2006
Cooking the transistors will indeed be more cost effective.

The problem with the lead soldering temperature is that you will not be able to limit the exposure to 10 seconds without forced cooling, which may affect the solder.

Though, as you said, it will probably take more than a slightly raised junction temperature to destroy the device. I'd try scoping it before and after, though, to check that you're hearing the performance of an uncooked device


Thermal conductivity should indeed be significantly better with a metal-metal bond. Keeping it thin can be accomplished with a vice while the solder is still in its liquid state.

Obviously, you'll have to make sure you don't end up shorting anything; I can't recall whether there's a live backplate on the thing.

I like the copper plate idea in that regard. Giving the device a larger effective area to dissipate from, and then isolating this area from the heatsink, neatly avoids some problems.
__________________
--
"Do not attempt to stop this cicular saw with your hands or genitalia." - Warning Label
  Reply With Quote
Old 16th September 2006, 11:14 AM   #1013
jkeny is offline jkeny  Ireland
Banned
 
Join Date: Feb 2007
Location: Dublin
Have you looked into the PC solution for CPU heatsink adhesive http://www.arcticsilver.com/as5.htm - don't know if the thermal conductivity is enough but it's good to > 180 degrees C.

Modern CPUs get damn hot so I imagine this along with maybe some water cooling would be cheaper than the Indium solder & no cooked JFets.

JOhn
  Reply With Quote
Old 16th September 2006, 02:38 PM   #1014
Sheldon is offline Sheldon  United States
diyAudio Member
 
Join Date: Dec 2003
Location: San Diego
If I try the Tix solder, I'll play with some proxy parts first to get a process down.

It's not clear from the data sheet whether the backplate is live or not, but I could ask in the GB thread. I would think that a live backplate would be better for thermal transfer.

Jkeny, the arcticsilver stuff looks interesting, but without a relative value it's hard to tell how it compares to a solder bond. More interesting maybe is the statement they make about this vis a vis conductive greases. Makes me wonder if the conductive greases wouldn't be even better.

Sheldon
  Reply With Quote
Old 16th September 2006, 10:04 PM   #1015
suiraMB is offline suiraMB  Norway
diyAudio Member
 
Join Date: Aug 2006
The Arctic Silver 5 actually looks like a fairly good candidate for heat sinking, and it's not particularly expensive either.

Modern CPUs don't necessarily get all that hot, since they are heavily heatsinked, but they do generate a lot of heat. In fact, we'll soon have to start using liquid sodium or liquid potassium for cooling, as the heat per surface area is getting close to what you find in a fission reactor.

It would be very interesting to compare this compound to the solder based approaches.
__________________
--
"Do not attempt to stop this cicular saw with your hands or genitalia." - Warning Label
  Reply With Quote
Old 17th September 2006, 09:18 AM   #1016
jkeny is offline jkeny  Ireland
Banned
 
Join Date: Feb 2007
Location: Dublin
Here are some measures of thermal grease effectiveness http://www.overclockers.com/articles662/

Don't know how this compares to solder based approaches but some measures given in the article suggest that
- there's not much difference between greases (in the overall scheme of the cooling problem)
- Greases are about 1/20 to 1/40 as effective as pure metal

However, this is assuming a perfect contact with metal - this is not the case - so we need greases to fill in microscopic uneveness of contact surfaces. How does this fact sit with the idea of a solder interface? Do we not still need some microscopic thermally conductive filler between surfaces?

Quote:
Thermal Conductivity of Selected Pure Metals: W/mK

* Silver: 417 W/m K
* Copper: 394 W/m K
* Gold: 291 W/m K
* Aluminum: 217 W/m K

You've probably seen this data before, so no surprise here. Now let's contrast that with the

Thermal Conductivity of Selected Thermal Greases: W/m K

* Arctic Silver III: >9.0 W/m K
* AOS Thermal Compounds 57000:7.21 W/m K
* Shin-Etsu G751: 4.5 W/m K
* AOS Thermal Compounds HTC-60: 2.51 W/m K
* Thermagon T-grease 412:1.3 W/m K
* Radio Shack Thermal Grease: 0.735 W/m K
  Reply With Quote
Old 17th September 2006, 02:47 PM   #1017
Sheldon is offline Sheldon  United States
diyAudio Member
 
Join Date: Dec 2003
Location: San Diego
A solder interface on a properly prepared cleaned and fluxed surface should fill in all spaces.

Sheldon
  Reply With Quote
Old 17th September 2006, 07:58 PM   #1018
suiraMB is offline suiraMB  Norway
diyAudio Member
 
Join Date: Aug 2006
As Sheldon said, the solder interface should work, because you keep it liquid while using a vice to press the device to the sink, so it should let the air out. The pressure on the vice will make the solder layer as thin as possible. I believe the indium solder can properly wet the surfaces. And the cleaning/fluxing should eliminate other impurities that could impede heat transfer.

Of course, the thermal conductivity of indium (81.8W/m/K) is lower than that of the heatsink, but I assume some of the other components of the solder have higher conductivity.

One approach that would be excellent with these devices, is to get a copper block with a water-maze and peltier element (something like this)
to enhance the cooling. And, of course, use solder to attach the device(s).

The theoretical minimum thermal resistance in this way is about 2.0K/W, due to the junction-case thermal resistance. Add a single peltier element, and you can lower the temperature by 20 deg C, which means you can get close to 45W continous dissipation.
__________________
--
"Do not attempt to stop this cicular saw with your hands or genitalia." - Warning Label
  Reply With Quote
Old 17th September 2006, 08:53 PM   #1019
jkeny is offline jkeny  Ireland
Banned
 
Join Date: Feb 2007
Location: Dublin
Hey SuiraMB - don't speak Norwegian - can you translate please.

Are these prices Norwegian Kroners? 1 Kr = .12

Has anybody actually succeeded in doing this indium soldering of Lovoltech devices to copper?

John
  Reply With Quote
Old 18th September 2006, 03:30 AM   #1020
suiraMB is offline suiraMB  Norway
diyAudio Member
 
Join Date: Aug 2006
Hey, John.

The rough and quick translation for the first one, which I assume is the one you're looking for, is as follows:

Quote:
http://www.mamut.com/homepages/Norway/1/5/oikas/subdet16.htm

OiK Pelt-Maze is a Danger Den Maze 2-1 water-block with a 172W peltier element factory mounted. The block itself is attached to the 4 holes surrounding the socket A and socket 423 (Pentium 4) CPU socket.

The block ships with 3/8 inch connectors (10mm). This is the best current option for those of you who wish to upgrade your water cooling equipment with the addition of a peltier element.

This block is one of the coolest (lit. rawest) you can get hold of for peltier cooling, the only better performer we know of being the direct-die-mounted "Genesis" block. The peltier element used here costs almost NOK 500 by itself. Using this block, you can expect full-load temperatures approaching the freezing point.

Bear in mind that the peltier element will require a 24V supply to perform optimally.

Price: NOK 1190,-
Also, NOK 500 is roughly USD 75 or EUR 60 or GBP 40, while NOK 1190 is roughly USD 180 or EUR 150 or GBP 100. Bear in mind that these prices probably include VAT, which is 25% in Norway.

Note also that I'm not endorsing this product, just giving an example of the approach.

With a copper router, three copper blocks, a peltier element, two or four O-rings, two nozzles, some screws and silicone, you could probably make something like this quite easily. And there are several companies that will supply something like this.

EDIT: As to the indium soldering bit, I know someone suggested it here, and I believe he succeeded in it. Either way, it shouldn't be all that hard. Indium itself has a melting point low enough that semiconductors that can take a 175degC junction temperature can be attached with the pure metal itself. The solders are compounds, so they melt at even lower temperatures (there are indium compounds that are liquid at room temperature).

EDIT2: Here is a picture of how you might do something like that, minus the peltier element. A pin matrix is probably simpler to make, though.
__________________
--
"Do not attempt to stop this cicular saw with your hands or genitalia." - Warning Label
  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



New To Site? Need Help?

All times are GMT. The time now is 07:41 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