defective SAD1024?

[Auke Haarsma]

I have a couple of SAD1024's. I am trying to verify them in a 9V electric mistress.

I have a couple of working SAD1024 (working=flanging from EM).

However some of the SAD1024 do not flange. Bias and clock have been adjusted if and as required per IC.

If I compare the 'good' IC's vs the 'bad' IC's I notice:
-input signal present at pin 2, 15
-output from pin 5, 12 is present with the good IC's. The 'bad' IC's do have a signal there too, but at a much lower volume (I can't hear it, I use the scope to see the signal). The signal is about 10 times lower than that of the good IC's.

Would this indicate a defective SAD1024?

Or, is there a way to bump up the output signal?

[Auke Haarsma]

anyone?

[Mark Hammer]

Is there any way to use only one half of it?  There was a period where Reticon marketed an SAD512 that was essentially a one-sided 1024, in which only one half of the stages (512) were guaranteed.

[Auke Haarsma]

They are stamped as SAD1024's.
If I am correct, both sides of the 'bad' IC's have a very low output. So using one side does not look feasible.

[R.G.]

Use your scope. If you have
(a) good signal going into the input pin
(b) good, solid ground-to-power supply clocks and non-overlapping phases
(c ) correct output loading

then a good SAD1024 will return a signal that's within 1db or so of the input, and stairstepped.

If you don't get the signal, the chip's bad.

But be very sure that item c is correct before tossing the chip.

[Auke Haarsma]

thanks Mark and RG.
in the 9v EM the outputs have a 470r via a 10k resistor to gnd. would that be proper loading?

http://gaussmarkov.net/layouts/mwelecmiss/mwelecmiss-schem.png

[StephenGiles]

What happens to the clock signals with the defective?? SAD 1024 plugged in? With the SAD unplugged, measure the voltage on pins 8 &10 (joined) and 3 & 14 (joined) - each should read about 4.5 volts. Now replace the SAD and measure the voltages again - they should be the same, if drastically different you probably have a defective SAD 1024.

[Auke Haarsma]

Thanks Stephen. The voltages are stable, with and without the sad1024 in there.