Thread Quality in Aluminium - Problems with the threads stripping out

[p1stonbroke]

Hi, I have a heat sink plate that is attached to a PWB using 6 x M2.5 x 6 screws. We are having problems with the threads stripping out, despite using considerably less force than might normally be expected (0.4 NM), inspection of the holes prior to thread insertion shows some variance, on one hole the threads might be fairly well defined (clear rigdes and troughs) whereas on another, they might be more rounded.
I am looking for a suggestion on how to determine what is actually acceptable and what is not. My company does not have spare funds to buy expensive visual measuring equipment, nor do we have the facilities to microsection to measure - I'd like a method that can be used by our goods inwards guys to say 'No' or 'Go' to each batch of heatsink plates as they come in.
Any suggestions?

[Jim Wynne]

Originally Posted by p1stonbroke

Hi, I have a heat sink plate that is attached to a PWB using 6 x M2.5 x 6 screws. We are having problems with the threads stripping out, despite using considerably less force than might normally be expected (0.4 NM), inspection of the holes prior to thread insertion shows some variance, on one hole the threads might be fairly well defined (clear rigdes and troughs) whereas on another, they might be more rounded.
I am looking for a suggestion on how to determine what is actually acceptable and what is not. My company does not have spare funds to buy expensive visual measuring equipment, nor do we have the facilities to microsection to measure - I'd like a method that can be used by our goods inwards guys to say 'No' or 'Go' to each batch of heatsink plates as they come in.
Any suggestions?

Under what conditions are the threads stripping? Are you sure that assemblers aren't cross-threading the holes? This is a common problem with threads in soft materials. You have to try and isolate the problem in order to find a solution. Here are some questions that might help lead you in the right direction:

• How are you inspecting the holes now? If you're using thread plug gages, try to find out if the thread condition you've visually identified is detectable with a thread gage (i.e., those with visually questionable threads might be looser or tighter on the gage). Mark some parts and see if they strip in production.
• Can the problems be associated with a single operator? I used to work in a job where teflon fittings with pipe threads were used. The fittings weren't cheap, and we kept getting rejections from the line with the cause being "bad threads." It turned out that one assembler was too aggressive in installing the fittings and kept cross-threading them, which was very easy to do with the soft material.
• Is thread class specified on your drawing? If the drawing just says M2.5x6, the tolerance class defaults to "medium" for metric threads. Have a look here for more information: Tolerancing of Screw Threads
• Unless something simple presents itself, you might have to do some cutting (sectioning of parts in order to measure the threads directly). I know you say you don't have the facilities to do it, but that just means that you might have to send some parts out to be measured. If there's a disagreement with the supplier, it's to your advantage to have ammunition from an independent source.
• Have you discussed this with the supplier? Is he aware of the problem?

[Bill Ryan]

Welcome to the posting side of the Cove

In addition to JSW's questions:

• Are the heat sinks a die casting?
• If so, how many cavities are cast in the mold?
• Also if so - Are the holes formed by a core in the mold or are they machined?
• Are these self tapping screws?

[Wes Bucey]

Bill asks pertinent questions, as does JSW.

When you say

My company does not have spare funds to buy expensive visual measuring equipment, nor do we have the facilities to microsection to measure - I'd like a method that can be used by our goods inwards guys to say 'No' or 'Go' to each batch of heatsink plates as they come in.
Any suggestions?

I have a few comments. Even something as simple and inexpensive as a jeweler's loupe with the proper light source will tell you almost everything YOU need to know about the female thread.

Depending on how the female threads are formed, you have several options.

If the alloy is too soft and the female threads easily deform from the screws, change to a harder alloy (hardly any difference in heat sink value.)

Are you sure the screws are consistent? sometimes the manufacturing process for screws can produce an out-of-round condition that turns the screw into a thread cutter. Not likely, but it is a remote possibility.

Do installers REALLY know how to install screws without stripping? (this is an insulting question, but must be asked.)

When you mention budget, you may not be aware many low tech, low cost intruments and tools for checking threads exist. Usually, the situation you describe does not need machine sectioning of the female threads to examine the thread condition. Accurate, inexpensive go/no go male thread gages are available which can be used in incoming inspection to confirm the female threads meet specs. This should be done by the supplier already. Do you get his inspection sheets?

Have you bothered to ask the supplier of the female threads to help you track the problem cause? Often this is the fastest and least expensive solution.

[RWest]

The only thing I think that wasn't mentioned is the minor diameter. You must determine the proper minor diameter and check (using a gage pin) that it is within spec. for the thread in question. The thread can pass with the plug go/not go gage and still have OOT minor diameter. The plug gage only checks the pitch diameter which is measured across the actual thread angle. I could go on but I think this is the most likley culpret barring the other things already mentioned.

Cheers,

R West

[MysterHK]

Originally Posted by p1stonbroke

Hi, I have a heat sink plate that is attached to a PWB using 6 x M2.5 x 6 screws. We are having problems with the threads stripping out, despite using considerably less force than might normally be expected (0.4 NM), inspection of the holes prior to thread insertion shows some variance, on one hole the threads might be fairly well defined (clear rigdes and troughs) whereas on another, they might be more rounded.
I am looking for a suggestion on how to determine what is actually acceptable and what is not. My company does not have spare funds to buy expensive visual measuring equipment, nor do we have the facilities to microsection to measure - I'd like a method that can be used by our goods inwards guys to say 'No' or 'Go' to each batch of heatsink plates as they come in.
Any suggestions?

p1stonbroke:

I am just throwing out an idea, but I was wondering if that aluminum part could be made with pressure-applied, self-clinching fasteners instead of threaded holes?

There's another idea I remembered back from my Navy days. Stripped holes would be tapped out and helicoils would be inserted. I don't know if there is a manufacturing equivalent to do this efficiently.

[Stijloor]

Originally Posted by p1stonbroke

Hi, I have a heat sink plate that is attached to a PWB using 6 x M2.5 x 6 screws. We are having problems with the threads stripping out, despite using considerably less force than might normally be expected (0.4 NM), inspection of the holes prior to thread insertion shows some variance, on one hole the threads might be fairly well defined (clear rigdes and troughs) whereas on another, they might be more rounded.
I am looking for a suggestion on how to determine what is actually acceptable and what is not. My company does not have spare funds to buy expensive visual measuring equipment, nor do we have the facilities to microsection to measure - I'd like a method that can be used by our goods inwards guys to say 'No' or 'Go' to each batch of heatsink plates as they come in.
Any suggestions?

Originally Posted by MysterHK

p1stonbroke:

I am just throwing out an idea, but I was wondering if that aluminum part could be made with pressure-applied, self-clinching fasteners instead of threaded holes?

There's another idea I remembered back from my Navy days. Stripped holes would be tapped out and helicoils would be inserted. I don't know if there is a manufacturing equivalent to do this in large quantities.

You are responding to an old post (30th June 2005, 07:09 AM) and I wonder if the Original Poster (OP) is still around...

However, for our Fellow Covers who are still interested in this issue, I found these...

Stijloor.

[MysterHK]

That is an awesome product, Stiljoor! Truly amazing!

I'm definitely going show this to my production manager tonight and maybe he can bring this up at the next engineering meeting.