Nelson Pass DIY F-5 take 4

Russellc, Why not try to get a group buy going in BT. I need a pair for a build I have coming up for another AKer.

Hey Jeff, I still have that pair you gave me in my closet. I was going to save them for a F1 build, but I don't see that happening anytime soon. You're welcome to them.
 
in researching my latest project , this F5 , i've gotten so many conflicting viewpoints . first off Nelson recommends a 6 x 8 x 2 heatsink for each output device with at least a base thickness of 1/4" which is 96sq ins / side .

from there i see the conrad MF35-151.5 which is @ 14 x 6 x <2 which is @ 84 but with shorter fins . conrad quoted me @ $170 for 2 shipped to mass usa .

heatsinkusa unfortunately only sells one extrusion suitable and it's @ 10' wide , however several folks have reported good results using 8" tall sections of it for their f5's . the fins are 2.5" .

i bought 4 sections of the heatsinkusa cut at 7" and am going to cut them down in width buy 2 " ea for 16 x 7 x 2.5

DSC_0241.jpg


these things are massive . almost 9 lbs ea .

DSC_0239.jpg


the four of these shipped was $144
Those sinks should work out well for you. I wish Heatsinks USA had a reference for thermal dissipation that was standard. Maybe I'm just reading it wrong. If you look through Nelson's article on the F5, the thermal dissipation target regarding the heatsinks for a standard build (no bias tweaking) is .6 deg C/watt.

I would be inclined to keep the heatsinks as they are. First, big is always better, lol. I have no problem with a high stature for an amplifier. Second, how are you going to cut the sinks? With a 10" circular blade, you'll be toward the maiximum of cut height (and probably the maximum torque on the saw). If you have a band saw, the only limiting thing is the length of the cut, it may be possible.
 
Last edited:
Those sinks should work out well for you. I wish Heatsinks USA had a reference for thermal dissipation that was standard. Maybe I'm just reading it wrong. If you look through Nelson's article on the F5, the thermal dissipation target regarding the heatsinks for a standard build (no bias tweaking) is .6 deg C/watt.

I would be inclined to keep the heatsinks as they are. First, big is always better, lol. I have no problem with a high stature for an amplifier. Second, how are you going to cut the sinks? With a 10" circular blade, you'll be toward the maiximum of cut height (and probably the maximum torque on the saw). If you have a band saw, the only limiting thing is the length of the cut, it may be possible.

Heatsink USA may be willing to do the cut if asked. He does business on EBAY a barrredboss or something like that, another member bought a length and had him cut in half I believe...but why not use the hole thing? (Other than a 20 inch deep amp.:D

I also wonder if one section would be enough for standard F-5? Nelson called for a 6 X 8 x 2 for each fet, this is two inches short. It could be made taller, but this is said not to help much. At least fins are 2.5 inches. Anyone out there used this piece for F-5?



Russellc
 
Last edited:
hey guys :) i guess if you have unlimited real estate for your components , dimensions don't matter . thing is i kinda do and don't because i've got too much stuff . i've got my av system and my music system and between both of them they take up alot of space especially considering i'm a vinyl junkie with 2 turntables and they get the top shelf .

i'm a cabinetmaker w/ a 5 hp table saw , but i'll probably cut them on the bandsaw and finish on beltsander . 20" deep is just way too much . i'm shooting for 16 " deep .

these are beautiful chunks of aluminum and his cuts are amazing .

over at the other place several people are using 2 8" tall pieces , one per channel successfully .

although the success of many of the builds are not backed up by real data . such as temp readings . however there are many people using sinks that are whimpy compared to these .

can't promise , but i might fire up one channel bolted to just one sink this weekend and my new dmm does temp readings . very curious myself .

cheers Woody
 
Hm-m-m-m-m, just got thru reading Anvila's post on the F-5 thread on DIYaudio, perhaps I will follow his lead on this one and use the 2.0 Cviller board and use dual pairs of Toshiba outputs.....he claims increased bass among other things.

The board also allows for cascoding, which in turn would allow for higher B+
rails.....hm-m-mm-.

Or maybe I'll just proceed as planned, I'm ordering multiple heatsinks and could do that for number 3.

Another thought is to mount various board versions to heat sinks, which could then be swapped out board and all....Most of the F series seems to use about the same power supply.:scratch2:

Russellc

Hey Russell, I went looking for that user post (Anvila) to no avail...could you provide the post link for the dual pair outputs...:thmbsp:

Thanks, Jon
 
Hey Russell, I went looking for that user post (Anvila) to no avail...could you provide the post link for the dual pair outputs...:thmbsp:

Thanks, Jon

It was posted in the monster F-5 thread just a few days ago. Its not a "build"
thread, he's just telling about his dual pair toshiba F-5, a post or two.

See threads: 10425-10442 or so....

Russellc
 
Last edited:
hey guys :) i guess if you have unlimited real estate for your components , dimensions don't matter . thing is i kinda do and don't because i've got too much stuff . i've got my av system and my music system and between both of them they take up alot of space especially considering i'm a vinyl junkie with 2 turntables and they get the top shelf .

i'm a cabinetmaker w/ a 5 hp table saw , but i'll probably cut them on the bandsaw and finish on beltsander . 20" deep is just way too much . i'm shooting for 16 " deep .

these are beautiful chunks of aluminum and his cuts are amazing .

over at the other place several people are using 2 8" tall pieces , one per channel successfully .

although the success of many of the builds are not backed up by real data . such as temp readings . however there are many people using sinks that are whimpy compared to these .

can't promise , but i might fire up one channel bolted to just one sink this weekend and my new dmm does temp readings . very curious myself .

cheers Woody
Your living space is, of course, going to rule. I kind of understood the cutting was necesary for your situation.

I've read nearly all of the threads at DIYaudio about the F5 (dropped off on the original thread at some point) and have looked at what others have been doing for heatsinks, mostly because I wasn't encountering the excessive heat problems I read about.
I came away with the thought that a lot of builders aren't using heatsinks that are up to the task. I'm in no way referring to your proposed plan and the Heatsink USA sinks as being inadequate. From the pictures and what I've read, they look to be the shiz.

I think your idea about measurements would be very good. If we can compile some real-use numbers to the thermal game, it would benefit all. I did do some measuring of the F5 I have but have lost the data. In the spirit of all this, I believe I'll get some better measuring gear and have at it, again. If you note your temps and some others can do the same, at least a comparison of different sinks can be made.
 
Your living space is, of course, going to rule. I kind of understood the cutting was necesary for your situation.

I've read nearly all of the threads at DIYaudio about the F5 (dropped off on the original thread at some point) and have looked at what others have been doing for heatsinks, mostly because I wasn't encountering the excessive heat problems I read about.
I came away with the thought that a lot of builders aren't using heatsinks that are up to the task. I'm in no way referring to your proposed plan and the Heatsink USA sinks as being inadequate. From the pictures and what I've read, they look to be the shiz.

I think your idea about measurements would be very good. If we can compile some real-use numbers to the thermal game, it would benefit all. I did do some measuring of the F5 I have but have lost the data. In the spirit of all this, I believe I'll get some better measuring gear and have at it, again. If you note your temps and some others can do the same, at least a comparison of different sinks can be made.

I started a "Heatsinkusa inquirey" sort of thread on DIY and have had some good responses. One knowledgeable member has a post there that shows a computer simulation of the sink and temps and so forth. It indicates using only one of these per side results in the mosfets running 60+ degrees.

Another member measured a dual pair F-5 on a single sink, very hot. He mentioned he may try it with a single pair just so the matter will be resolved. I hope he does so!

http://www.diyaudio.com/forums/pass-labs/184899-f-5-heatsink-query.html


http://www.diyaudio.com/forums/pass-labs/184508-finished-f5.html#post2493407
Russellc
 
Last edited:
Russell, how's your helper from post #243 doing?

Not even a year old and almost 12 lbs. He's a household favorite and loved by all.

Russellc

Then, and now.
 

Attachments

  • IMG_2730.JPG
    IMG_2730.JPG
    128.6 KB · Views: 13
  • IMG_2717.jpg
    IMG_2717.jpg
    70.7 KB · Views: 8
  • IMG_3205.JPG
    IMG_3205.JPG
    108.3 KB · Views: 17
Last edited:
hey guys :)

i decided to give it a real world test . not just a simulation . using just ONE chunk of heatsinkusa 10.08" x 7" and mounting one channel , BOTH mosfets .

DSC_0244.jpg


it's been up and running for almost an hour . bias has stabilized at @.58-59v and it's nice and warm . temp readings pretty much everywhere on the heatsink are only 37-38 C . at the bolt and washer holding the fet they measure @62 c .

granted this is without an enclosed case , however , i'd say 38 C readings leave quite a bit of leeway .

cheers , Woody
 
k , awesome corned beef (burp) , it's been running close to 4 hours and after it stabilized i biased it at .60v (it floats a lil ) .

DSC_0250.jpg


thats degrees C on the meter taken @ 2" over the fet . i could leave my hand on it all day .

the mosfets are mounted with silicon heatsink compound and keystone micas .

the input is shorted .

am i missing something ?

cheers Woody
 
Thanks, for posting that, Woody. Just for clarification, it looks like you are using the C Viller Ver.2 boards, both the amp boards and the PS supply board. What transformer are you using and which output devices? Also, can you share what you used to measure?

FWIW, your sink temps are within a few degrees of what I was measuring, IIRC. I will post some results of the M&M sinks in the future.
 
hey westend :) the amp boards are the cvillar (really nice chap) designed boards that are being sold by diyaudio store . the power supply boards i got directly from cvillar .

the trannie is an antec 5218 500 18 18 .

the meter is a bk precision 2706b http://search.digikey.com/scripts/dksearch/dksus.dll?Cat=2294086&k=2706b

i bought it because it has the temp feature and also the ability to measure cap up to 20,000uf . really pleased with it and the backlite is really sweet .

:)

oh and outputs are irf , i bought the kits from http://www.tech-diy.com/Store/CloneKits.htm
 
Last edited:
Yes, but the heat of the sink doesnt tell the whole story. A lot depends on the heat transfer of the insulator under the mosfet. The heat sink could be reasonable, while the mosfet itself overheated in a worst case scenario.

I cant remember for sure, but I thought the limit was around 50C. (For the mosfet) I dont know if the part you measured is reflective of the heat of the mosfet, but if I'm remembering my reading, 62C would b too hot for the MOSFET, and I dont know if the mosfet IS, in your case.

One member on DIYAudio forum showed simulations using this sink, resulting in "borderline" heat measurments, but thought using Kerafol/Keratherm might make enough difference.

Can anyone add to the maximum mosfet juncture heat and so forth? I'm about at the limit of my knowledge here. Sure looks like a substantial hunk of sink!

Thanks for trying this with this heatsink, there are a lot of people wanting to see results.

Russellc
 
Last edited:
read replies over there for answer . it says for those not in the loop

from a mod there

"no, its probably not hot enough....if measurements are reliable

place your hand on the heatsink, to check the 'meter'

'could' be related to heat transfer issues
but I think mosfet would be a lot hotter if so "

and my reply not to him but to russell :

50 C is for the heatsink as far as i know according to NP :

"Again, the heat sinks on this amplifier run fairly hot, and you want to make sure
that they get adequate ventilation. They will run at around 25 degrees C.
above the ambient temperature, which puts them around 50 degrees in the
average listening room. At this temperature you should be able to put your
hand on them for about 5 to 10 seconds or so."

also "Heat Sinking
At 1.3 amps per channel, you will see idle heat dissipation of 62 watts. To keep the
temperature rise of the heat sink to 20 deg C." NP

temps mentioned are heatsink temps , not at the mosfet . my mosfets are mounted using NP's methods as far as i can deduct .

so far everyone here seems to claim this or that . at one point i said that it would be nice if folks could post ACTUAL temp readings for their specific setups and was immediately shot down by a mod here saying they were irrelevant .

these ARE actual temp readings for a heatsinkusa chunk of their 10.08" x 7" heatsink biased at .60v using an antec 5218 , cvillar/diyaudio boards .
 
Gotcha! If 50c is just for the sink, then that sounds like good news. The HeatsinkUSA extrusion is very affordable, and looks to be a nice sink.

I was thinking it was 50C for the mosfet, in this case there is a little more room, and Kerafol should also make a small difference here as well.

Very nice work, thanks for the efforts!

russellc
 
now BEFORE everyone runs out and buys a 10 x 7 from heatsinkusa !!!!!!!!!!!!!!!!!! remember this is a test situation . once encapsulated w/o enough ventilation ?

i did this to prove a point and my voltage and temp readings WERE real .

i bought 4 chunks of this because someone there said i needed more , and someone there ran a simulation and ....

then someone there said a 6" chunk would probably be enough so i just had to ...

hey i like hanging out over here with you guys .

i'll be running some more tests this week and probably have a working amp by weeks end .

i just HATE when people tell me i can't do something :)

cheers Woody
 
Back
Top Bottom