Deep hole tolerance costs

[TokyoChris]

I'm a design engineer at a company in Japan and I need to design a plate in which to insert a heating element. The heater is cylindrical with a 4mm OD (+0, -0.08), and the plate is 170mm long in the direction of the heater. The manufacturer's recommendation is a hole machining clearance of "0.05 or less (one side)".

I have only limited experience, so I might just be being overzealous. I figure the hole should be as tight as possible, so as close to 4mm (+0, -0.08) as possible. I suggested 4mm H7 but my boss/supervisor insisted that it's too precise and said to just go with 4.2 (+0.05, -0.05) which is outside of the manufacturer's recommendation. Given the 170mm length of the hole, I guess it'll be gun-drilled - maybe from both ends. Concentricity isn't strictly required, but there is other geometry at each end, so a deviation of less than 0.1mm is desired.

Would that stricter tolerance really be so difficult, or is it that much more cost-effective to have a hole that size? I work in Tokyo so we run by JIS, but only loosely (it appears).

Regards,
Chris

 

[SwinnyGG]

Much depends on exactly what, in your mind, your stated tolerances mean.

If all you care about is that the diameter of your hole is within those limits, then I don't think this is too big of a deal.

If what you actually need is for the inside profile of the hole to be round, cylindrical, and perpendicular to within that window, this gets more challenging.

Another way of asking this is: how flexible is your heating element, how critical is it that the space around it is a consistent/precise air gap, and how critical is it that the wall thickness of this hole is consistent/precise?

 

[racookpe1978]

No. Your heater needs to expand and contract as each heat cycle starts and stops.
Dirt, corrosion, gum, crud are going to get in the hole and on the heater as well - it has to go in and out.
And, this is a very small (but long) little heater: Assume the heater is not perfectly straight - either before insertion or after. When you insert it, it rubs against one side. So it touches along one wall, then is forced against the other, then back against the first wall. You can't force it - the heater will break off in the hole and ruin that hole. Or force expensive "Drill it out again" surgery. Which then gives you a worn, scratched or overdrilled manual hole. Or one with a bad heater in it that you don't know has failed.

A very tiny hole is incredibly expensive (difficult) to drill to minute tolerances. And, if it is not drilled exactly to specification, what are you going to do? Throw out the whole plate and start over?

 

[EdStainless]

It is also customary to fill these holes when you insert the heater to assist in thermal coupling.
MgO is commonly used as a paste.
This lets you use holes that are little looser and still get good heating.
What material is the plate? How much expansion mismatch will there be?
I would have gone to 4.1mm +/-0.05 as a first try to see what stack-up I get.
This is going to be an expensive hole.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[MintJulep]

I'd suggest changing the approach entirely.

Go for a groove and a bolted cover. Possibly with Bellville washers to deal with thermal expansion.

 

[TokyoChris]

Thanks for the replies, guys.

how flexible is your heating element, how critical is it that the space around it is a consistent/precise air gap, and how critical is it that the wall thickness of this hole is consistent/precise?

The heater is 304SS and they haven't specified any other flexibility other than the dimensional tolerance. The plate is an aluminium alloy used to transfer heat. Presumably the vertical variation shouldn't have too much influence on the heating itself.
The recommended clearance gap was specified by the manufacturer. That's my only real guideline.
The point of the hole is to fix the heater, which is there to transfer heat. Given that, the ideal situation is the hole having as much surface area in contact with the heater as possible.

No. Your heater needs to expand and contract as each heat cycle starts and stops.
...
A very tiny hole is incredibly expensive (difficult) to drill to minute tolerances. And, if it is not drilled exactly to specification, what are you going to do? Throw out the whole plate and start over?

From my hand calcs (scribbles), the heater should only expand as much as 0.01mm up to 150C. The plate will also expand, but that will increase the force between the surfaces and increase heat transfer, right?
Either way, we don't do this machining in-house. My job is just to give a feasible design with accurate dimensions and tolerance. Incorrect machining is another company's concern as far as I can imagine.

It is also customary to fill these holes when you insert the heater to assist in thermal coupling.
MgO is commonly used as a paste.
This lets you use holes that are little looser and still get good heating.
What material is the plate? How much expansion mismatch will there be?
I would have gone to 4.1mm +/-0.05 as a first try to see what stack-up I get.
This is going to be an expensive hole.

We will use a thermal grease to fill the hole. The grease likely only has a fraction of the thermal conductivity of the plate, so I'd rather not rely too loosely on that.
The plate is A6061. If my calcs are right, that should expand only 0.045mm. If we sum the expansions, we get about 0.05mm or 0.06mm (not accounting for expansion in other directions due to obstructions).
I considered 4.1mm +/-0.05 but then that would also be outside of the manufacturer's recommendation. I guess the real question is why the manufacturer recommended such a near clearance if it's that difficult to machine. This heater isn't even their narrowest, nor their longest.

Go for a groove and a bolted cover. Possibly with Bellville washers to deal with thermal expansion.

My company's existing design, prior to my arrival, was a slotted semi-cylindrical groove cut a 2mm or so into the plate, then covered by a sheet metal fixing. I gave my boss an alternate design of more or less the same thing but with a male-shaped fixture to fill the gap tightly, bolted through the plate to fix from the top. (see my cross-section ascii art below).
My boss preferred the through hole design for it's simplicity and fewer parts. To get a similar clearance, wouldn't that design end up being more or less the same complication as the hole design? Granted the level-ness is less of an issue, but extra parts to be milled with a ball-tip to a tight tolerance (and obstructing the heat transfer).
It does seem like a good idea. Any other thoughts?
_________
|--O--|
_________

 

[3DDave]

The hole gets bigger, right? I can't tell if your summing of the expansions is correct; 0.045 - 0.01 = 0.035.

 

[TokyoChris]

The hole gets bigger, right? I can't tell if your summing of the expansions is correct; 0.045 - 0.01 = 0.035.

You're right - should've subtracted instead of adding. Either way, the expansion would be difficult to accurately predict. So long as the heater doesn't expand drastically faster than the plate hole, it should be of no real concern. The same problem would occur in the alternate design.

 

[CouplingGuru]

We have to make holes of that nature periodically for hydraulic relief ports. Granted our application isn't as tolerance critical, but because it is low volume we just ED bore them. It isn't too bad in price. If you are making 10,000 a year, may not be the way to go. But if you are making a few hundred a year probably the best bet, and you can get your H7 tolerances.

When it comes to couplings we are always here to help.

http://WWW.PSCCOUPLINGS.COM

 

[TokyoChris]

we just ED bore them

Thanks for the direct advice! ED bore? What's the ED? I guess the cost limit is up to my boss. I figured that a slight cost increase for higher efficiency should at least be considered.

 

[btrueblood]

Electrical Discharge Machining, EDM - in making holes, it would be ED boring.

 

[Compositepro]

Cartridge heaters are very common and almost every one gets placed in a hole. The sheath of the heater will be far hotter than the aluminum plate you are heating. When there is a gap, heat transfer will be poor, so the sheath will get hotter and expand until it contacts the aluminum (within limits). The cartridge is as flexible as any drill bit, so getting jammed because the hole is not straight is a non-issue.

The only reason to put grease in the hole would be to prevent corrosion. Read the application notes in the Chromalox or Watlow catalogs. They explain this.

 

[TokyoChris]

Electrical Discharge Machining

Thanks. I'll pose it to my boss when he's back, but I think he's pretty set on just gun-drilling.

Cartridge heaters are very common and almost every one gets placed in a hole. The sheath of the heater will be far hotter than the aluminum plate you are heating. When there is a gap, heat transfer will be poor, so the sheath will get hotter and expand until it contacts the aluminum (within limits). The cartridge is as flexible as any drill bit, so getting jammed because the hole is not straight is a non-issue.

Thanks, mate. From my hand calcs, this heater should only expand 90microns or so up to 150C, which wouldn't reach the 4.1mm hole even at the manufacturer's upper diameter tolerance limit (let alone once the aluminium also expands). That's assuming the heater is 304SS throughout.

Am I being too overzealous by wanting to stay within the manufacturer's recommendations? Has anyone had such cartridge heater holes manufactured and got any tips for tolerance clearances?