Tektronix input attenuators

[John_BS]

I'm posting this in case anyone else comes across a similar problem.

I've just been calibrating a 7A26, which employs the Tek modular thick-film attenuators, which are arranged in this unit (reading from the input) 10x, 10x 5x, 2x . The nominal input sensistivity is 10mV/div with no attenuators in circuit.

Channel 2 aligns fine. Both channels are spot-on in terms of actual LF attenuation across the range. On channel 1, (square-wave input) everything is OK untill we reach the "check" on 500mV/div (i.e the 10x & 5x are in circuit), where we get about 5% undershoot on the extreme leading edge. Similarly when the 5x pops in again at 5V/div. If the 5x is tweaked to correctly compensate 5V & 500mV, it overshoots when on 50mV/div.

I checked all the components around the input and output, and cleaned all the switch contacts. No change. Swapped a spare 5x attenuator into place: no change. Similarly the preceding 10x.

Eventualy I found the cure was to spray the 5x module with contact grease, as the trimmer c's were quite "loose", and this presumably fixed what I can only conclude was a high-resistance rotating joint on one of the trimmers.

This topology is used extensively in 4xx and 7000 series scopes.

[ronbryan]

John

I have had trouble with a 5X attenuator thick film module on a 7A26, but it turned out to be a dry joint where one of the module pins joined the thin film circuit.

Do you have any idea why the spare 5X attenuator module did not fix it?

Ron

[John_BS]

Ron:

The more complete history is as follows:

The original 5x attenator was almost unsettable: as soon as the trimmer touched the HF adjuster, the response would leap all over the place.

The 5x from an old donor module displayed the problem outlined above, and the original was apparently fixed with the contact grease. So I suppose I ought to try the grease on the donor sample before drawing any firm conclusions.

John