[volt-nuts] Junior Member Lost Shirt on 720a KVM Purchase

Tony vnuts at toneh.demon.co.uk
Fri Mar 27 11:32:37 EDT 2015


You could consider making your own - it's a switch, not rocket science. 
Layout a radial contact pattern and get some PCB's made up at a cheap 
PCB facility for $10 or so. Get them gold plated - possibly $20 or so? 
Ideally the gold would be alloyed to make it harder and more durable and 
2 or 3 um thick. Make the wiping contacts from gold plated copper strip 
and fashion some sort of spring to apply an appropriate contact force. 
Use several if necessary to reduce the contact resistance sufficiently. 
Solder to the inner most part of the radial contacts to minimise thermal 
EMFs (by keeping all solder joints as close as possible for best thermal 
equilibrium).

With respect to thermal EMFs it should outperform those expensive 
switches as:

a) It would be all copper - those Dallas suggested have silver-alloy and 
brass contacts with copper terminals and the Elma datasheet doesn't even 
specify the materials (ridiculous for such an expensive part).

b) The fixed and moving contact solder joints can be significantly 
closer together, around the shaft rather than around the periphery, and 
thus easier to minimize the temperature differences.

Indexing can be achieved in lots of ways including drilling holes in the 
PCB and having one of the moving contacts, or a spring loaded ball 
bearing of an appropriate size, to partially drop into the holes. 
Alternatively, create notches around the circumference which a shaped 
metal strip drops into under spring pressure.

The main concern for this construction might be leakage across the PCB 
and durability depending on how well you can control the contact force. 
With a bit more complexity you could even arrange for the indexing 
mechanism to raise the moving contacts (to reduce, not eliminate wiping 
action) in between switching positions.

If leakage is a concern you could even bolt pie shaped radial segments 
cut from gold or silver plated copper sheet to a PTFE disc. That would 
have the additional advantage that both the fixed and moving contacts 
could be made from the same piece of copper, reducing thermal EMFs even 
further. The contact area, most of each pie segment, can be quite large 
compared to those at the circumference of commercial switches with the 
potential for very low contact resistance.

Finally, use thin strips of the same copper for any interconnections 
rather than copper wire of slightly different composition to eliminate 
anther dissimilar metal to metal joint - at one end anyway.

Must be worth considering given the high prices being asked for the 
commercial switches.

Tony H

On 26/03/2015 21:51, Stan Katz wrote:
> Hi Dallas and Frank,
>
> Dallas, the diameter of the switch you pointed me to is indeed too large.
> It's only available at Digikey for a very affordable $92.00USD. The shaft
> is also not compatible with the 720a construction.
>
> Frank, I called up Fluke for a quote on the switch. Delivery in one week
> for only $1060.00USD!!!! I'm still waiting for a quote from IET.  The ELNA
> 04 doesn't look like a candidate, but I have requested a quote anyway.
>
> Shame on Fluke for using an outside supplier for that faulty switch. This
> costly, and impractical to repair 720a has taken away all of the thrill of
> classical analog metrology for me.
>
> I think that I should consider a used Fluke 5440b and give up on lash ups
> needing the 720a.
>
> Stan
>
>



More information about the volt-nuts mailing list