Complete Novice Needs Help

Status
Not open for further replies.
Thanks Nar ,

You have a really good point ... in that , if I cant rely on the tolerance of the silicone to be as closely matched as the fets ,using matched fets is pointless ..

BTTDB = back to the drawing board ....

Ill just stick to the first idea of using wire .... or this ?
 

Attachments

  • DSCN0156.jpeg
    DSCN0156.jpeg
    21.6 KB · Views: 157
Thanks Nar ,
You have a really good point ... in that , if I cant rely on the tolerance of the silicone to be as closely matched as the fets ,using matched fets is pointless ..

BTTDB = back to the drawing board ....

Ill just stick to the first idea of using wire .... or this ?

I think you got it wrong. I'm sorry and I apologize as English is not my native so I still do some horrible mistakes 🙂

I was just wandering, not guiding you to the drawing board 🙂 Come on 🙂

It's not the tolerance of the silicone that I was pointing out. Once you'll block it between the metal bar and the mosfets, then applying torque it will be tight. I was just wandering if the silicone would be as stiff/thick once it would have reached roughly thermal operating point. If it goes a little softer it might loose the stiffness it has at ambient, and thus loosing the applying strength of the mosfets onto the metal sink.

Once the "absolutely best" method for clamping has been described, I think one more hole wouldn't hurt to get your work done nicely 🙂

Not really off topic : I always tried to drill then taper the holes in the raw aluminium. However it's not the best idea as the metal itself is not very hard and after some unmount/mount of the devices, it can happen that the screw doesn't hold the fet anymore ... so thermal coupling can be ruined 🙁

I wonder if someone already used inserts ? metallic inserts so you drill the sink at a higher gauge, then hammer down in a metallic insert that is already tapered ... maybe even wood inserts could work ... Any opinions ? Just a hint though


Best,


nAr
 
No need to be sorry

Nar ,

I appreciate the input. This is my first project. The silicone layer may or may not work ! I have spent too much money and time on this project already . I will KISS as the say get it done and running and think about my next project !

I guess drilling more holes is possible . The better the contact the higher I can crank the bias . Ill have to wait till I get the ally bar . I will just use silpads between the back of the fet and the bar .

You talked about the tapping and drilling .... I would use a thermal epoxy if a repair was needed of that kind 🙂

BTW,I had a look at your cables .... beautiful work sir !

Rich
 
Nar ,

I appreciate the input. This is my first project. The silicone layer may or may not work ! I have spent too much money and time on this project already . I will KISS as the say get it done and running and think about my next project !

I guess drilling more holes is possible . The better the contact the higher I can crank the bias . Ill have to wait till I get the ally bar . I will just use silpads between the back of the fet and the bar .

You talked about the tapping and drilling .... I would use a thermal epoxy if a repair was needed of that kind 🙂

BTW,I had a look at your cables .... beautiful work sir !

Rich

Thanks for your input, however subjective cable sound research is for now stopped - no money and extra time to finance that as I also have to make a living - that doesn't exclude the possibility to make high quality DIY interconnects for me 🙂

And yes, I have spent many hours ( for free ) on the website as I created almost all graphics 🙂

Best,

nAr
 
Sure Steve,

If I stripped a thread ,whichI have done in the past .

1, drill out hole , at least .5 mm larger than the size you need .

2, Tap thread at the new hole size .

3, clean out hole with acetone = nail varnish remover ( dont get caught by the Mrs during PMS )

This step is very important . I use a syringe to force the acetone in to the hole . It will evaporate leaving surface degreased and ready to glue.

4, mix epoxy and some metal powder. I use brass powder that I got from an art shop .
I have been working with resins in various applications for many years and Devcon
is my prefered

ITW Devcon

5, lightly coat olive oil ,or any light oil on the threads and wipe thread with a cloth ,
just wipe ,dont rub .

6, fill hole with resin using a large syringe . You can get them from the drug store .

7, place your new screw or bolt in the hole , tread the screw in , if you just push it
in your will get voids.Let it set , a warm lace is best . Resins cure better with
a little heat .

8, remove screw and finish flush to existing material .

9, I always lubricate all threads when fixing the screw for the final time .


Rich
 
Last edited:
Sure Steve,

If I stripped a thread ,whichI have done in the past .

1, drill out hole , at least .5 mm larger than the size you need .

2, Tap thread at the new hole size .

3, clean out hole with acetone = nail varnish remover ( dont get caught by the Mrs during PMS )

This step is very important . I use a syringe to force the acetone in to the hole . It will evaporate leaving surface degreased and ready to glue.

4, mix epoxy and some metal powder. I use brass powder that I got from an art shop .
I have been working with resins in various applications for many years and Devcon
is my prefered

ITW Devcon

5, lightly coat olive oil ,or any light oil on the threads and wipe thread with a cloth ,
just wipe ,dont rub .

6, fill hole with resin using a large syringe . You can get them from the drug store .

7, place your new screw or bolt in the hole , tread the screw in , if you just push it
in your will get voids.Let it set , a warm lace is best . Resins cure better with
a little heat .

8, remove screw and finish flush to existing material .

9, I always lubricate all threads when fixing the screw for the final time .


Rich

Thanks for the tip 😎 really awsome, technic but requires the adequate elements 😱

Best,

nAr
 
Have you tried Helicoil inserts. Possibly easier and more reliable solution for stripped threads.

Sizes M3 and up.

Cost effective solution for me despite sourcing to Australia from a UK supplier on eBay.

Nar, these screw in rather than hammer in.
 
Last edited:
Have you tried Helicoil inserts. Possibly easier and more reliable solution for stripped threads.

Sizes M3 and up.

Cost effective solution for me despite sourcing to Australia from a UK supplier on eBay.

Nar, these screw in rather than hammer in.

Those seem a better alternative to tapering into thick aluminum sinks ... !!!
Do you have a link to the UK supplier ? Thanks ! 🙂

nAr
 
I will just use silpads between the back of the fet and the bar .
you don't need an extra "silpad" between the clamp bar and the FET case. Silpads and similar are conformal. They change shape at elevated temperatures. Basically they start to "flow" by going into a plastic stage to enable them to fill all the little crevices/scratches etc.
This will release the force of the clamp bar.

You need springiness/compliance that is as near perfect "elastic" as you can get. Some materials "creep" with time and/or with stress. These are no good either for trying to maintain a near constant pushing force.

In the first instance just clamp the bar down onto the bare FET case. Some elastic deformation of the clamp bar and of the FET package will give a tiny bit of compliance. Even the bolts stretch slightly when tightened = more compliance.
 
Have you tried Helicoil inserts. Possibly easier and more reliable solution for stripped threads.
Helicoil are often used to "repair" damaged threads.
A better use for Helicoil is to make the original helicoiled for all tapped holes.
Then you can remove and replace studs/bolts with ease and almost guarantee never damaging the thread.
I have a couple of secondhand heatsinks that came with ALL the tapped holes helicoiled. It must have cost a lot to buy first time around !
 
Thermal!

Hi Steve,

all materials expand and contract at different rates when exposed to temp fluctuations . Straight epoxy would come loose in time . Thermally conductive adhesive would allow the heat to migrate at the pretty much the same rate as the surrounding materials.

I hope this explains a little .

Rich
 
Hi Jimbo,

helicoil is a good system , but I have never had a faliure with the method I described .
My method is based on what I carry in my workshop for such problems .

I work with all kinds of materials . Some epoxy is Lloyds registered . It has to be insurable. Devcon R&D is in bed with NASA ... and they dont make O rings .

Rich
 
Status
Not open for further replies.