Stolen Trademark Amplifier from Jim's Audio on EBAY - Page 37 - 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 2013, 12:32 AM   #361
spurlte is offline spurlte  United States
diyAudio Member
 
Join Date: Aug 2011
Default Heat sinks

I have a table saw with carbide tip saw blade and as long as I feed it very, very and very slowly it cut beautifully. Anything else I used overheated the aluminum and it just puddled onto the teeth of the tool. I envy you guys for finding your fan type heat sinks. Damn!!!! And already TO-3 drilled. I have these hugh heat sinks where the fins run parallel lengthwise. 14 inches long, 6 inches wide and 2 inches deep. I was even thinking to cut across the fins to create spikes, put two together faces each other and place fan, push-pull at each each. I got these very cheap at a metal scrap yard. Kaplaars, I was wondering when you very going to run your tunnels verticle. As Superman says, "Up, up and away!"
Attached Images
File Type: jpg heatsinks.jpg (393.3 KB, 246 views)
  Reply With Quote
Old 10th January 2013, 02:59 AM   #362
spurlte is offline spurlte  United States
diyAudio Member
 
Join Date: Aug 2011
Quote:
Originally Posted by Kaplaars View Post
Hi David,

I have to say, in all honesty, that I do not share the observation Fix has made. I actually can't hear any difference when experimenting with bias values ranging from 100 mV and up. I begin to experience slight deteriation in sonic quality when I dial the bias far below 100 mV.

But I have to say, I don't listen music at very loud volumes. So the amp has not to deliver lots of power to drive my speakers. That is why, in my case, a lower bias will do just fine. When your speakers draw more power than your bias is set, and you exceed this value, then you can experience more distortion due to crossover distortion, but below it should not occur.

But cooling is very, very important. Lot's op people underestimate it. I for example did haha. That is why I want to add even bigger tunnel coolers because 1) I am bored haha, 2) because this way the fans have to displace even less air to get the same results and 3) because I offcourse want to go all the way ;-). After some evaluation I think it is after all beter to separate the tunnel coolers afterall, and place them upright just as fix did and the original KSA-100 has.

If I am correct the original KSA-100 uses tunnel coolers which are 12cm long. If I am not mistaken Fix is using tunnel coolers which are about the same length. Even then I think the KSA-100 runs pretty hot. So you will need probably more cooling. For my new plans I have bought these beauties (29cm long each, almost twice as heavy as the ones I use right now):

Click the image to open in full size.

I am thinking of cutting them through at about 18CM and place them upright. Does anybody has some experiencing with cutting aluminum profiles like these?

Liquid cooling would be very verry cool David! I would love to see that. Even better, it does not produce a lot of noise. I think for me that would be a little to expensive. I think I stick to aircooling.
Yes air cooled. But attaching a plank, to attached T0-3s, to a sealed rectangular conduit with a tube attached to a liquid circulator attached to an automobile heater core with a small fan then a tube back to the rectangular conduit. Just a thought, but that is available for computer CPUs on small smaller scale. The heater core may even not need a fan. There would be even liquid traveling--could add a dry sump canister of maybe a pint or so. I used this in my dry sump oil cooler unit in my race car. Crankshaft do not like oil sloshing in the engine's oil pan and most expensive--Porsche--use an oil radiator cooler. So they figured why take the oil storage out of the oil pan (away from the heat too) and put a canister mounted aside the inner engine compartment. Now, Kaplaars, or anyone. Would the output transistors NEED to be at atleast a optimal operating temperature?
  Reply With Quote
Old 10th January 2013, 03:11 AM   #363
spurlte is offline spurlte  United States
diyAudio Member
 
Join Date: Aug 2011
Default Cutting aluminum

Quote:
Originally Posted by spurlte View Post
Yes air cooled. But attaching a plank, to attached T0-3s, to a sealed rectangular conduit with a tube attached to a liquid circulator attached to an automobile heater core with a small fan then a tube back to the rectangular conduit. Just a thought, but that is available for computer CPUs on small smaller scale. The heater core may even not need a fan. There would be even liquid traveling--could add a dry sump canister of maybe a pint or so. I used this in my dry sump oil cooler unit in my race car. Crankshaft do not like oil sloshing in the engine's oil pan and most expensive--Porsche--use an oil radiator cooler. So they figured why take the oil storage out of the oil pan (away from the heat too) and put a canister mounted aside the inner engine compartment. Now, Kaplaars, or anyone. Would the output transistors NEED to be at atleast a optimal operating temperature?
Kaplaars. I was very successful with cutting my aluminum sheets and bars up to 1 quarter inch .250. Like I said before, I used my table saw feeding it to the carbide tipped 10 inch saw blade. Even when I used my drill bench, the aluminum melted, but with the table saw, it sent very small chips out just like wood chips with no melting. You can cut one side, turn it over and cut the other side. If you have a table saw.
  Reply With Quote
Old 2nd February 2013, 10:05 PM   #364
diyAudio Member
 
Kaplaars's Avatar
 
Join Date: Jul 2010
Location: Wageningen
Hi Friends!

I've been thinking about the overhaul of my KSA when spring break arrives. Main upgrade will be adding other tunnel coolers, together with bigger (14cm) fans. Now, my cooling tunnels have nice pre-drilled holes for 32 TO-3 devices. I will use one tunnel and cut it in halve. This would mean I could double the amount of output devices if I want to. Just for curiosity, would this be beneficial?

A benefit which comes with doubling the amount of output devices will be that every TO-3 device has to dissipate halve of the heat it is dissipating right now. If I am correct the total heat generated will be the same, because overall bias is still the same, thus also heat generated. But chances of getting out of SOA will be smaller, because the individual devices have to do less work.

A disadvantage could be that when doubling the amount TO-3 devices, chances of instability will rise because of internal ifferences between the devices (deviation in bias per device). Probably using 1 ohm emitter resistors will help to tackle this problem. Another problem could be oscillation.

What is wisdom? Am I forgetting something ? I like the sound right now, and I have always learned 'don't fix if it aint broken', but maybe improvement could be made :-).

Also I am curious about optimal orientation of the coolers themself. What would be the best orientation? Fans on top and pushing air through the tunnels down or reversed thus pulling the air through the tunnels instead. An other orientation could be placing the fans beneath and pushing air upright through the tunnel, or go for pulling the air through instead? And offcourse I can use the push / pull principle like I am using right now, so one fan on top, and one fan beneath.

Sorry for the overflood of questions, but I am very curious :-).
__________________
Michael van den Born - PA4MW
  Reply With Quote
Old 3rd February 2013, 04:14 AM   #365
spurlte is offline spurlte  United States
diyAudio Member
 
Join Date: Aug 2011
Default Devices

Quote:
Originally Posted by Kaplaars View Post
Hi Friends!

I've been thinking about the overhaul of my KSA when spring break arrives. Main upgrade will be adding other tunnel coolers, together with bigger (14cm) fans. Now, my cooling tunnels have nice pre-drilled holes for 32 TO-3 devices. I will use one tunnel and cut it in halve. This would mean I could double the amount of output devices if I want to. Just for curiosity, would this be beneficial?

A benefit which comes with doubling the amount of output devices will be that every TO-3 device has to dissipate halve of the heat it is dissipating right now. If I am correct the total heat generated will be the same, because overall bias is still the same, thus also heat generated. But chances of getting out of SOA will be smaller, because the individual devices have to do less work.

A disadvantage could be that when doubling the amount TO-3 devices, chances of instability will rise because of internal ifferences between the devices (deviation in bias per device). Probably using 1 ohm emitter resistors will help to tackle this problem. Another problem could be oscillation.

What is wisdom? Am I forgetting something ? I like the sound right now, and I have always learned 'don't fix if it aint broken', but maybe improvement could be made :-).

Also I am curious about optimal orientation of the coolers themself. What would be the best orientation? Fans on top and pushing air through the tunnels down or reversed thus pulling the air through the tunnels instead. An other orientation could be placing the fans beneath and pushing air upright through the tunnel, or go for pulling the air through instead? And offcourse I can use the push / pull principle like I am using right now, so one fan on top, and one fan beneath.

Sorry for the overflood of questions, but I am very curious :-).
I know I am just a novice, but is it possible to run a POT to adjust resistance in series with the 1 Ohm emitter resistor to adjust internal differences between devices? Sorry about answering a question with a question.
  Reply With Quote
Old 3rd February 2013, 04:55 AM   #366
diyAudio Member
 
Stuart Easson's Avatar
 
Join Date: Oct 2002
Location: Santa Cruz, CA
Send a message via MSN to Stuart Easson Send a message via Yahoo to Stuart Easson
I've said this in the other thread, so I'll be brief:

The easiest way to ensure good matching between the output transistors current is to use the largest reasonable value resistors that are matched. I used resistors that were nominally 5%, but better than that in practice, and I measured the current for all 6 pairs of output transistors which were not matched, save by being from the same lot.

Getting resistors that match to 1% is cheap and easy, and if you really want to do better than 1% you can get there pretty easily by buying more and measuring them with a decent meter. However, shooting for better than 5% matching is going to be more or less pointless unless you also make sure all the pcb tracks and solder joints in line are also similarly matched. It is also rendered pointless unless you can keep the thermal tracking across the entire heatsink equally perfect. Why? a 1c differential in the temperature of the transistors will induce a ~3% difference in the Vbe, ie the hotter transistor will try and steal more current at about 3% per degree C, the larger the emitter resistor, the larger it's induced voltage to counteract the drop in Vbe.

An important thing to note: The more output transistors you put in parallel the larger the emitter resistors need to be: Adjust the value of the resistor so the overall paralleled emitter resistance matches the original design. "But why?" I hear you ask, well the value of the output paralleled emitter resistors determines the bias voltage to achieve a given emitter current but it also sets the bias in the driver stage, ie lower bias voltage to get the output stage into class A reduces the current in the driver stage...If either half of the driver stage runs out of current, it's partner is still working to drive the output but as a pair they are no longer in class A and the amp will not sound like a Krell amp, even if the output stage never leaves class A.

For 6 pairs of output transistors we used 2ohm resistors, whose parallel resistance matches the ksa50's 2 output pairs 0.66 pretty well.

Ok, that was not brief...sorry

Stuart
  Reply With Quote
Old 3rd February 2013, 04:54 PM   #367
AVWERK is offline AVWERK  United States
diyAudio Member
 
Join Date: Jan 2004
Location: So Calif.
Thinking about these issues with heat and emmiters and such.... At the CES in January this year, Vandersteen,s new amp coming out soon has liquid cooling, no emmiter resistors at all mounted on copper. Push,push instead of push pull Sankens. The biggest R core tranny I have ever seen sitting down on the bottom of the pyramid shaped enclosure with 6np2/6h30 tudes driving the whole thing.
Just refreshingly different and probably very nice sounding considering his track record.

Regards
David
  Reply With Quote
Old 9th February 2013, 05:10 PM   #368
neychi is offline neychi  Croatia
diyAudio Member
 
Join Date: Feb 2005
Location: zagreb
Hi Guys,

Well, it seems I have a problem. When I turn the amp on without the output stage, for testing, in first few seconds everything looks fine, but then, the drivers (mje15032/33) and the bias transistor (mje15032) goes in smoke and they have shorted B-C-E. Output shows then 48V (rail voltage). Everything else seems to be ok. I checked everything and I don't have a clue. Bias trimmer potentiometer is at 1/10 of the value (5k). Any idea is welcomed.
__________________
Kind Regards
  Reply With Quote
Old 9th February 2013, 07:44 PM   #369
xslavic is offline xslavic  Moldova
diyAudio Member
 
Join Date: Feb 2011
maybe its the zvp transistors polarity,if you have replacements check it again and of course all diodes on the pcb
  Reply With Quote
Old 14th February 2013, 06:19 AM   #370
neychi is offline neychi  Croatia
diyAudio Member
 
Join Date: Feb 2005
Location: zagreb
It's working!!!

Toshiba 2sa968/2sc2238 were obviously a fakes. They shows normal readings, but when they are connected to the voltage, something goes wrong. Anyway, I put 2sa1011/2sc2344 instead and everything is good now.
__________________
Kind Regards
  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
Jim's Audio F5 power supply board Hikari1 Pass Labs 35 4th December 2013 11:41 PM
Jim's Audio Goldmund Telos 390 clone Tyimo Solid State 9 5th March 2013 03:28 AM
Jim's Audio F5 referenced kit. HiVi Guy Pass Labs 21 22nd November 2012 11:24 AM
Stolen items? kctess5 Multi-Way 14 23rd February 2012 11:02 PM


New To Site? Need Help?

All times are GMT. The time now is 11:13 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