Kovar vs 1215 1

[GSwitch]

We are a leading manufacturer of motion, acceleration and impact switches. Typically, these devices are built in transistor packages (TO-18, TO-5, etc)Some time back, someone decided to switch from Kovar to C1215 for headers, in an effort to reduce costs. However, we use nickel cans and have encountered weld spatter problems (resistance welding, Benchmark welders). Since these devices must be absolutely free of contamination, loose particulate spatter is a serious problem.

I am trying to convince management that returning to Kovar is the best solution, as its low resistivity and molten flow characteristics are clearly better than C1215. Comments or suggestions from those with greater experience with Kovar/nickel welding would be of great use.

Thanks,

GSwitch

 

[metman]

Apparently the headers require machining, otherwise why would they switch to C1215? My guess is that the Phosphorus and Sulfer in the 1215 are causing the spatter. The gain in machinability they have seen from Kovar to C1215 is probably substantial. Therefore maybe a compromise is in order. 303 S.S. is a free machining grade and would be a compromise for the machining and welding aspects, but I don't know about resistivity since I do not have Kovar info handy.

Hope this helps Jesus is The TRUTH

 

[GSwitch]

I have data handy for Kovar wire, where Specific Resistivity is rated at 49 micro-ohms/cm.

Specific Resistance is rated at 294 ohms/CMF.

My regards,

GSwitch

 

[TVP]

If welding is a requirement, SAE 1215 steel is a terrible choice. I am not surprised at all that you are experiencing problems with this grade, as metman mentioned, the high levels of P & S which aid machinability, are extremely detrimental to weldability.

Before making any suggestions about returning to Kovar, alternative materials, etc., perhaps you could help clear up some confusion I have on the application. What components are being welded (thickness, geometry, material, etc.)? Why was Kovar used in the first place? Anything else that would be helpful?

 

[GSwitch]

This switch (an FAA certified crash sensor) is built in a modified TO-5 package. Typically, inside diameter of the nickel can is .314", with a flange diameter of about .355-.365". The header inserts into the can and is resistive-welded around the diameter using weld tools that support the flange and squeeze it and the header flange as the welder cycles. The weld itself is made in an inert atmosphere as the switch is backfilled with 90% N2 and 10% He. This is done to enable us to locate switches with a hermetic seal leak rate greater than 1x10-6. Those that have a greater leakage rate are rejected.

Kovar is the usual material for header construction. You can view a standard extended TO-5 package at:

http://www.aerodyne-controls.com/a3.htm

Most of our headers are made of Kovar. This particular header is machined, and in the interest of cost cutting, 1215 was adopted. We haven't experienced any serious issues with glass bonding to the 1215, but Kovar does bond better as demonstrated by a higher percentage of leakers with 1215.
All headers are plated, usually with gold, but sometimes with silver per customer requirement.

So, we need the following attributes:
1) Ease of machining.
2) Low resistivity.
3) Good bonding with glass.
4) Readily accepts nickel strike and plating.
5) Weld compatibility with nickel can.

Is there anything else that I need provide?

My regards,

GSwitch

 

[GSwitch]

One more item, flange thickness of the header is .020". Thickness of the mating can flange is .012".

My regards,

GSwitch

 

[TVP]

GSwitch,

Thanks for the additional info, especially the schematic, it was quite useful. Now I understand why Kovar was originally used-- the CTE is the same as that of borosilicate glass, thereby minimizing thermal stresses in the metal-to-glass bond. Plus, it will result in a good weld with the nickel can.

I would like to reiterate that SAE 1215 steel is a terrible choice when it comes to weldability, and in this application, it is even worse, since you are welding to a nickel alloy can, not to an iron-base material. I fully understand the desire to lower the cost, and 1215 will result in both material savings and in lower machining costs. However, I would strongly recommend against using this grade, for a variety of reasons, including poor weldability and different CTE than nickel or kovar or glass. The best quality option is to revert to Kovar.

You might investigate a better machining grade of a nickel-based alloy. One example is Alloy R-405 (Monel R-405), which is similar to Alloy 400 (Monel 400), but with S added for better machinability. This would hopefully reduce the cost penalty somewhat vs. SAE 1215, but not be as bad as Kovar. I don't want to comment too much on glass bonding, since I do not regularly work with nickel alloys. You will need to investigate this further on your own.

I recommend you check out the following links to review more information on Alloy R-405. Special Metals is the leading supplier of nickel-based alloys, so they will probably be able to help you the most. Allegheny Ludlum (

http://www.alleghenyludlum.com)

is another producer that may offer this alloy.

http://www.specialmetals.com/publication/tech_bulletin_r-405.pdf

http://www.espimetals.com/tech/machiningnickel&alloys.pdf

 

[GSwitch]

Thank you very much for your time and trouble.

I will be investigating R-405 immediately.

Sincerely,

GSwitch

Project Engineer,
Aerodyne Controls