# comparator input latches high?

I am using a a TI TLV3202 comparator (U2B) with a bridge circuit as shown in the schematic below.

It is followed by a DG4157 analog switch. This is related to this question I had asked earlier.

After power-up and when RLS is approx. 50 k$\Omega$ range things seem to work - the voltage at TP13 varies with RLS. However, if RLS goes to approx. 1M or is open circuited, the voltage at TP13 goes to approx. 5V (as expected), but then STAYS at 5V, regardless of the resistance of RLS. In other words, it seems that the positive input of the comparator latches 'high'.

What could be causing this?? The lead length of the positive input is relatively long (~3 in) - could this be some sort of parasitic capacitance feedback effect from the output? If so, is there a way to address it without laying out the board again?

• Is your DG4157 in SC-70 or miniQFN (they have different pinouts)? What is the voltage at TP1? You need to better localize your issue. Unless you have verified that TP1 is at the voltage you expect, the issue may not be the comparator at all. – Connor Wolf Sep 23 '13 at 6:59
• @ConnorWolf - The DG4157 is the SC-70 package. The voltage at TP1 varies with RLS as I would expect (in fact that's how I know RLS is varying). – Westerley Sep 23 '13 at 13:48
• Do you get the same behavior without the analog switch in the circuit? How about if you break the connection between TP1 and the junction of C19 and D1? – Joe Hass Sep 23 '13 at 15:21
• @JoeHass Taking the DG4157 out of circuit results in the comparator+bridge to work exactly as it should. So, what you and ConnorWolf stated/implied seems to be correct - the problem is not the comparator per se. What could be happening in the DG4157 to affect the comparator circuit? I've expanded the schematic to show what's on the other side of the DG4157. The voltage at TP12 is about 100mV with the DG4157 OUT of circuit. – Westerley Sep 23 '13 at 17:31