Fluke 12 Wrong reading in AC

[Spectrum]

Hi guys,
I bought this DMM, small size, fast reading. It is quite accurate in CC and Resistance, but it gets wrong readings in VAC. They are approximately halved in the low rates and greater in the higher rates. There are not many components and I have tested almost all of them, without finding significant differences. There remains only the pin 38 of U1, the one who receives the signal after the resistive network Z1.
Does anyone know this device and is aware of any suggestions? 

[Digorion]

Hello!
Service manual might be useful: https://elektrotanya.com/fluke_10_11_12_multimeter_sm.pdf/download.html
Are you sure pin 38 and not 39?

[Spectrum]

Yes sorry, pin 39...

[retiredcaps]

It is quite accurate in CC and Resistance, but it gets wrong readings in VAC. They are approximately halved in the low rates and greater in the higher rates.

Examples of the above please wrt to VAC.

I only have wall outlet for VAC source.  What do you get when you measure your wall outlet?

[retiredcaps]

It is quite accurate in CC and Resistance

What is CC?  Do you mean DC volts?

[Spectrum]

Yes, excuse me I did a mistake...CC=DC, Corrente Continua(in italian)=Direct Current
I did not test the wall outlet, probably this value is quite exact. I have a Voltage Calibrator and under 20 VAC the values are about half (not exactly but about) so I can read 8-9 VAC instead 20VAC, 3-4-5 VAC instead 10 VAC, 1 VAC instead 2 VAC, 0 VAC instead 1 VAC. Above 20 VAC, the values are higher, for example 28 VAC instead 25 VAC, 55 VAC instead 50 VAC, 105 instead 100 VAC...probably the value of the wall outlet is acceptable but I did not measure this value...

[bdunham7]

If Z1 and C14 all test good, it looks like the problem would be internal to U1, so not likely fixable.  Maybe test the S1 contacts just for good measure.  If the DC ranges all work well, it isn't likely to be anything in the ranging part of Z1 either.

[Spectrum]

Ja, I think this too...

Next time I unplug pin 39 and check if the exact value reaches that pin. If is so, the problem is U1

Any idea where to find it?

[bdunham7]

Ja, I think this too...

Next time I unplug pin 39 and check if the exact value reaches that pin. If is so, the problem is U1

Any idea where to find it?

The voltage won't reach pin 39 because most of it is dropped across the 10M input resistor part of Z1.  Then, presumably, an analog switch internal to U1 connects it to APV1-4, depending on range, which determines how much of the input the voltage sensing circuitry in U1 sees--presumably 1/10, 1/100, 1/1000, etc. 

As to where to find U1, it is almost certainly a proprietary chip that is long out of production.  Another Fluke 12 that has been physically damaged would be the only likely donor I can think of.

[Spectrum]

I know...
Or some warehouse fund. Some time ago I bought on ebay a box with many Fluke spare parts (flex, LCD, elastomery, pad, network res.) but no uP

[Spectrum]

I found a Fluke 10 board...assembled in my Fluke 12, it works like a Fluke 12 (Memory, Hold, Capacitance, Volt alert etc)

[retiredcaps]

I found a Fluke 10 board...assembled in my Fluke 12, it works like a Fluke 12 (Memory, Hold, Capacitance, Volt alert etc)

This thread is 10 years old, hard to believe I participated in it, but it's relevant to your comment about a Fluke 10 board working like a Fluke 12.  It has pictures and everything.

https://www.eevblog.com/forum/chat/fluke-25-my-first-digital-multimeter/

[Spectrum]

Yes, very interesting...exactly the same. One would think that all the specifications on the datasheet of the various models Fluke 1x are all fake