Surface Finishing

[julianthegypsy]

I definitely think it's a sound idea. If you do get round to trying it, keep us updated on how it fares, best method for application etc.

Will do Matt. I had a friend years ago who worked with gold leaf. I seem to remember he rubbed a brush to get some static on it to pick the leaf up then carefully painted it on so I'll start there. Off to evilbay to buy some leaf now and will let you know in due course :-)

 

[MattF]

I have white spirit but for some reason thought you had to use something special...and expensive, obviously :-)

That's what they try to convince you of so that they can flog you the citrus solvents and the like. I've just spent so many years removing that stuff that I've literally just tried every solvent I had to hand when I run out of whichever thing I had for it at the time, and especially when trying to get it out of clothes, that citrus stuff stinks for yonks if you apply it to material.

Off to evilbay to buy some leaf now and will let you know in due course :-)

Sweet.

 

[Rocketman]

3 things hit. Firstly you could just buy the aluminium ground on 2 faces if you want it really flat (less than 0.01 out over surface parallel/thickness). Secondly if heat transfer is so important you could water cool and just mill it flat and use Viton 'o' rings. Thirdly what confidence have you in the other surface being flat/parallel.

 

[julianthegypsy]

3 things hit. Firstly you could just buy the aluminium ground on 2 faces if you want it really flat (less than 0.01 out over surface parallel/thickness). Secondly if heat transfer is so important you could water cool and just mill it flat and use Viton 'o' rings. Thirdly what confidence have you in the other surface being flat/parallel.

I could buy ground aluminium, but at a huge cost. I'm making the blocks so the face is only one part of the equation.
As for water cooling, that's also on the plan, but without going into detail, I still need flat faces, again this is about the interface, it doesn't matter how cold the block is if its not getting the contact with the hot component. As for the other face being flat, well I'm making that too, hence the original question, and I'm fairly confident. :-)

 

[KemppiFrog]

.If I was lapping both components together, I wonder if I could make some kind of mount to spin one against the other and save some time and elbow grease? Once I get going, I should have rather more than three to do!

Not so sure about that idea, as the different speeds across the diameter would affect wear rate. They might fit each other, but not be flat. Glass is the way, handy to true up cylinder heads if no machinery around.
Sounds an interesting project, any more details?
KF

 

[julianthegypsy]

Not so sure about that idea, as the different speeds across the diameter would affect wear rate. They might fit each other, but not be flat. Glass is the way, handy to true up cylinder heads if no machinery around.
Sounds an interesting project, any more details?
KF

Yes, it would probably need something orbital, I was just trying to find a lazy way:-)
It is an interesting little project and I'll put some details up when it's ready. I did a lot of work on it a couple of years ago, got it ready to sell then got utterly shafted by the foundry who were making my castings. It took me ages to recover but this time I'm doing it all myself and using machined blocks. I still have enquiries from last time round but don't want to talk about it much until I have stuff ready to go, I'm sure you'll understand why.
I have two other little problems to solve and I'm just about there though. I need a reservoir that will also act as a heat sink. I've seen a hollow aluminium extrusion with fins all around but can't for the life of me find any for sale. The reservoir needs to hold one or two litres max.
The other little problem is connecting 8mm od copper pipe to an aluminium block. I can drill and tapwhatever is needed but I can't figure out what connections are best and what I should be drilling into the block. It sounds really dumb, but starting from scratch it's just so confusing trying to find the right thing!

 

[lshigham]

Yes, it would probably need something orbital, I was just trying to find a lazy way:-)
It is an interesting little project and I'll put some details up when it's ready. I did a lot of work on it a couple of years ago, got it ready to sell then got utterly shafted by the foundry who were making my castings. It took me ages to recover but this time I'm doing it all myself and using machined blocks. I still have enquiries from last time round but don't want to talk about it much until I have stuff ready to go, I'm sure you'll understand why.
I have two other little problems to solve and I'm just about there though. I need a reservoir that will also act as a heat sink. I've seen a hollow aluminium extrusion with fins all around but can't for the life of me find any for sale. The reservoir needs to hold one or two litres max.
The other little problem is connecting 8mm od copper pipe to an aluminium block. I can drill and tapwhatever is needed but I can't figure out what connections are best and what I should be drilling into the block. It sounds really dumb, but starting from scratch it's just so confusing trying to find the right thing!

One way to do it, not necessairly the best but it would work, would be to tap the block 1/4'' bsp and use a brass compression fitting like this http://www.ebay.co.uk/itm/8mm-x-1-4...DIY_Materials_Plumbing_MJ&hash=item1c3c0c2dfa

 

[julianthegypsy]

One way to do it, not necessairly the best but it would work, would be to tap the block 1/4'' bsp and use a brass compression fitting like this http://www.ebay.co.uk/itm/8mm-x-1-4...DIY_Materials_Plumbing_MJ&hash=item1c3c0c2dfa

That looks pretty much like what I want to achieve, my worry is how to ensure it doesn't leak around the bsp thread (I've had this problem before). The system doesn't have any pressure in it, but drips from the block are really irritating and silicon sealant is messy, loks bad and is very hit and miss!

 

[Paul99]

That looks pretty much like what I want to achieve, my worry is how to ensure it doesn't leak around the bsp thread (I've had this problem before). The system doesn't have any pressure in it, but drips from the block are really irritating and silicon sealant is messy, loks bad and is very hit and miss!

PTFE thread sealant paste (or even tape) would be better than silicon sealant. The other option maybe a dowty washer between block / adapter as per hydraulic couplings (if they are made that small).

 

[brightspark]

That looks pretty much like what I want to achieve, my worry is how to ensure it doesn't leak around the bsp thread (I've had this problem before). The system doesn't have any pressure in it, but drips from the block are really irritating and silicon sealant is messy, loks bad and is very hit and miss!

drilling and tapping it bsp will work fine its well proven even under great pressure . just use a decent jointing compound and not silicone. as most think its the answer to everything . when its all finished just do a pressure test on it if your not confident it wont hold

 

[Pete.]

Are you using plastic pipe? You can get nice push-fit connectors that screw into a bsp/npt thread.

 

[Shox Dr]

If you search for ali extrusions you'll find the ribbed sections, I know this as I was looking for something else the other day, and came across them. Are you using copper pipe so as to loose more heat? If not then plastic push fit pipe is available in many small bore sizes, search for Pneumatic Pipe fittings



Sounds interesting

 

[Robotstar5]

Used something like this as a manifold to run multiple air cylinders on this

 

[lshigham]

PTFE thread sealant paste (or even tape) would be better than silicon sealant. The other option maybe a dowty washer between block / adapter as per hydraulic couplings (if they are made that small).

Yes, dowty's are readily available in 1/4'' BSP, we have them at work. That's what I'd use

 

[Rocketman]

I did a quick model on Solidworks this afternoon. I will send the bill later lol!!!!

But anyway my company does injection mould tool design (plus other mechanical design sometimes).

This is generally a typical method of joining up water cooling from two plates. The plates only need to be milled and bolted together. The Viton 'O' ring seals it. The water ways have tapered 1/4" bsp water plugs in the end. You only need to tap straight though and use a bit of PTFE tape on the plug. I have shown 1/4" hose fittings - you could go with quick release if you want a really good job.

Waterways shown are 10mm. The tap drill for plugs and water fitting for 1/4" bsp is 11.80.

 

[julianthegypsy]

Wow, thanks for all the input. I'll be using copper as it will be near a heat source at over 200 degrees. Plastic could cause problems. The couplins with dowty washers look perfect.
I was going to mill a U shaped pocket in a 25mm slab, bore and thread the holes for the couplings in the end, then make a simple flat lid, which I know will seal easily with any jointing compound. One of the reasons for doing it like this rather than boring and connecting holes, which I had considered, is that the holes would be quite long and there would be a lot of tapping for plugging up holes. I also want a flatter shape, and want to bore a hole right through the middle (through the centre of the U) to clamp the whole thing down. My little centec should handle milling the pocket quite easily, I used it to make a 10mm slot in a piece of 6mm steel plate yesterday and it munched through it quite happily. A 15mm deep pocket in aluminium should be quite easy, even if I have to do it in two hits.
Thanks ever so much for all the replies, it really is unbelievably helpful. When I have everything working, I will posted all the details and loads of pics!

 

[AndyStobbs]

Just to throw in my 2 pence. I find that boss white and hemp is a much better way to mate two threads than PTFE tape. Especially if the fit isn't very tight to begin with, ie. Two BSPP fittings

 

[MrFluffy]

First thing that sprang to mind when reading this, is nowhere have you defined how flat you need it. Lapping on a sheet unless you are very careful will produce a slightly domed block with the rocking action, but as stated by Pete, you can see this with engineers blue and a surface master (surface plate etc). But its dependant on how flat you need it if you need to worry to the nth degree.
I'd send the piece out for surface grinding if I wanted to save time and the application called for that level of flatness for final finish. Find someone set up to grind heads and blocks flat on aluminium engines, as they will have the required silicon carbide wheels already mounted. Worst part of the job for someone who knows what they are doing is getting it to stick the magnet

 

[julianthegypsy]

First thing that sprang to mind when reading this, is nowhere have you defined how flat you need it. Lapping on a sheet unless you are very careful will produce a slightly domed block with the rocking action, but as stated by Pete, you can see this with engineers blue and a surface master (surface plate etc). But its dependant on how flat you need it if you need to worry to the nth degree.
I'd send the piece out for surface grinding if I wanted to save time and the application called for that level of flatness for final finish. Find someone set up to grind heads and blocks flat on aluminium engines, as they will have the required silicon carbide wheels already mounted. Worst part of the job for someone who knows what they are doing is getting it to stick the magnet

Should be within 0.001" per foot, There is no set standard for the surface finish, but I'll be going for as good as is practical. Was planning on using crocus paper.

 

[MrFluffy]

I think with that tolerance you are way beyond a sheet of plate glass or sandpaper taped down on something or similar methods. Ive seen some specs that plate glass is about 0.001" per INCH as it comes out, and that is assuming it is laid on a dead flat surface which supports it evenly (ie a surface plate) as it distorts to the shape of the supporting surface as glass is a deformable liquid at that level. Maybe I think more like a proper cast iron lapping plate with diamond compound and a lot of careful work against a master. With regards to grinding it, your going to have to find someone with a blanchard grinder and ask them to work to your tolerance, could be expensive. A accurate surface grinder will work to in the order of 0.002" per ft max and clamping forces will play hell with that figure.

Most the places I can find for heat transfer surface flatness for heatsinks are calling out a tolerance for surface flatness of 0.004" per inch. That'll be £££ instead of ££££££ territory though. Can't you just make the contact area bigger to allow a looser tolerance?