Surface Finishing

[graffian]

Just bookmarked one. Annoyingly there are two on ebay for just under £60 from allendale, but they have a 10mm shank. I habe a 1/4, 3/8 and 1/2" collets and I know already that I can't get 10mm in my collet, so I have to try and find a 10mm collet on a MT2 with a 3/8 whitworth thread. Oh joy...

Try making a dial gauge clamp to go around your spindle. I was thinking of buying one of those indicators but really wanted something I could clamp up without removing the tooling. Here's a pic of
it when I was boring my forks, It's made from scrap[an old bit of chain and a shovel pushrod;-)].

I was wondering why the OP was using copper leaf rather than silver? Or a squidge of heat sink goo.

 

[Lawnmowerboy]

Do you have a Jacobs chuck for your mil?

 

[julianthegypsy]

Try making a dial gauge clamp to go around your spindle. I was thinking of buying one of those indicators but really wanted something I could clamp up without removing the tooling. Here's a pic of
it when I was boring my forks, It's made from scrap[an old bit of chain and a shovel pushrod;-)].

I was wondering why the OP was using copper leaf rather than silver? Or a squidge of heat sink goo.

That looks possible, but surely ou can't rotate it?
I'm using copper because it should conduct better than heat sink goo. As it happens, I have bought some silver too :-)

 

[julianthegypsy]

Do you have a Jacobs chuck for your mil?

I have one but it doesn't have a thread for a drawbar. I suppose it would do for holding a shank though, thanks for that, I hadn't thought of it.

 

[Big Pete]

copper leaf sandwiched between may well make things worse. I used to be a thermal analysis engineer, the rule of thumb was that conductivity across a joint was 10% of the lowest material conductivity. That rule was based around experimental evidence and was generally reasonably reliable. It varies a bit with joint pressure, and a bit with surface finish (less so), and heat sink compounds (usually silver based) are the usual solution.

 

[julianthegypsy]

copper leaf sandwiched between may well make things worse. I used to be a thermal analysis engineer, the rule of thumb was that conductivity across a joint was 10% of the lowest material conductivity. That rule was based around experimental evidence and was generally reasonably reliable. It varies a bit with joint pressure, and a bit with surface finish (less so), and heat sink compounds (usually silver based) are the usual solution.

I've used heat sink compound and it doesn't seem that hot. All the ones I've seen are silicon based, whether or not they contain silver, and have to be applied perfectly to work, not too much and not too little, which is not ideal since getting it exactly right is a matter of guesswork and luck. Silicon really isn't very conductive. The idea is to replace the heatsink compound, not supplement it. Hopefully on a torqued joint, it should compress and fill the microscopic voids.

 

[Big Pete]

proper silver heat sink goo helps with the voids, the partcle size is absolutely miniscule, IIRC the stuff I used to see was a suspension in alcohol, which evaporated. Unless you have a huge amount of bolts, spaced evenly across the surface, and run them at ~80% of yield you aren't going to do much to the copper, you just can't get enough pressure, and the ally will likely move first. TBH, heat sink type devices aren't great, which is why fluid based cooling systems are so common.

 

[graffian]

That looks possible, but surely ou can't rotate it?

Yes it rotates, it's clamped on the spindle not the quill.

 

[julianthegypsy]

proper silver heat sink goo helps with the voids, the partcle size is absolutely miniscule, IIRC the stuff I used to see was a suspension in alcohol, which evaporated. Unless you have a huge amount of bolts, spaced evenly across the surface, and run them at ~80% of yield you aren't going to do much to the copper, you just can't get enough pressure, and the ally will likely move first. TBH, heat sink type devices aren't great, which is why fluid based cooling systems are so common.

Haha, it is a fluid based system, though for my own reasons it will have a heatsink on the cooling block too. The copper leaf really is thin, about 0.003". I'd be interested to know what brand of goo you used because as far as I've been able to ascertain, the difference between the them all is so small that there is no significant difference between the best and the worst.

 

[brightspark]

just out of curiosity. how have u tested and measured the heat transfer between the various materials and compounds and what were the results

 

[Big Pete]

I've no idea what the goo was, It was aerospace stuff and wasn't cheap, I wasn't doing the application either. 3 thou is massively thick compared to surface finish irregularities.

We measured the heat transfer rate using thermocouples on actual structures, IIRC the lab guys had a fancy setup that reduced the effects of radiation losses and such and used specially made sample pieces to control the heat flux. I think they used thermocouples for all the high end stuff as well. We were looking at temperatures up to ~1200C. Can't really give much in the way of details of results I'm afraid, in no small part to the fact that there was reams of it, and it was also (still is) commercially sensitive.

 

[brightspark]

I've no idea what the goo was, It was aerospace stuff and wasn't cheap, I wasn't doing the application either. 3 thou is massively thick compared to surface finish irregularities.

We measured the heat transfer rate using thermocouples on actual structures, IIRC the lab guys had a fancy setup that reduced the effects of radiation losses and such and used specially made sample pieces to control the heat flux. I think they used thermocouples for all the high end stuff as well. We were looking at temperatures up to ~1200C. Can't really give much in the way of details of results I'm afraid, in no small part to the fact that there was reams of it, and it was also (still is) commercially sensitive.

thanks for the reply but i was asking julian what conclusions he had come to and what were his measurments in relation to the various methods of heat sink transfer ??

 

[julianthegypsy]

just out of curiosity. how have u tested and measured the heat transfer between the various materials and compounds and what were the results

Very crude really, but I compare voltage outputs off the system. Can't really tell you any more...yet, but it does highlight the differences between what works and what doesn't very nicely.

 

[julianthegypsy]

I've no idea what the goo was, It was aerospace stuff and wasn't cheap, I wasn't doing the application either. 3 thou is massively thick compared to surface finish irregularities.

We measured the heat transfer rate using thermocouples on actual structures, IIRC the lab guys had a fancy setup that reduced the effects of radiation losses and such and used specially made sample pieces to control the heat flux. I think they used thermocouples for all the high end stuff as well. We were looking at temperatures up to ~1200C. Can't really give much in the way of details of results I'm afraid, in no small part to the fact that there was reams of it, and it was also (still is) commercially sensitive.

1200 degrees is way hotter than I'm going, but if the goo was aerospace I shall go hunting aerospace goo and see if I can find anything useful.

 

[brightspark]

Very crude really, but I compare voltage outputs off the system. Can't really tell you any more...yet, but it does highlight the differences between what works and what doesn't very nicely.

themocouples ??

 

[julianthegypsy]

themocouples ??

Haha. Hundreds of them :-)

 

[MrFluffy]

Try making a dial gauge clamp to go around your spindle. I was thinking of buying one of those indicators but really wanted something I could clamp up without removing the tooling. Here's a pic of
it when I was boring my forks, It's made from scrap[an old bit of chain and a shovel pushrod;-)].

I was wondering why the OP was using copper leaf rather than silver? Or a squidge of heat sink goo.

I do that but with a indicator off a mag base with a vee shape on its base. You can shift the chinese one and get a bad reading if your clumsy but I have a eclipse one thats donkeys years old that is brilliant for it.