Fluke 8808A not measure

[eduaqa]

I bought this multimeter as part, and I'm trying to fix
Initially when powering on, it was beeping, beeping and display was come on and off fast
Doing a initial investigation I notice that the 7805 marked was warming ... I removed it and test the rail that it supply voltage and was short circuited
I discovered to IC (latches) marked on the picture in short ... I replace the 2 IC and the rail now is having around 24ohms resistance, which still seems having some type of short circuit, because is cause a consumption of 370mAH, what seems too much, because when compared with the others rail which read very high resistance

When I turn it on now, the display works, the push button on the front change as should, however none of the measures works
I noticed that the 7805 is warming, as it should because it's providing 370mAH, also I notice that doesn't matter what function I choose I don't hear relays "clicking" and relays are also warming ... also measuring the voltage on the relays S and R coil that have around 4V, on both ... seems strange, I understood that should not have voltage on both at the same time, and maybe even not continuous

Any help are welcome

[Harry_22]

Hi!

Did I understand correctly that the software part of the multimeter works well now and all modes switch correctly. But you don't see any measurements and the relays don't click.
It feels like the measuring part itself is not controlled and seems to be missing.

Could you attach a photo in good resolution so that the names of the chips can be seen?

[eduaqa]

I'm adding some more pictures which has text of notes on them.
I replace the MOSFET that activate the relays coils but just realize that probably I used wrong MOSFET, because 1 side of the coil goes direct to the +5 rail, the other side of the coil goes to the source of the mosfet (yes strange because normally goes to the drain, I think is call high side). This assuming that the device that I noted is a MOSFET, than the drain goes to 0v ... in this scenario assuming it's a N channel (what I think I assumed wrong) the MOSFET will acts as a diode and relay will always have at least around 5v on the coil
Look the picture where I noted as "what device is this?". I assumed a nchannel MOSFET but seems that is not, because of the explanation above

[eduaqa]

Some pictures ... and more, it's no allowing me at onc3

[eduaqa]

More pictures with notes

[eduaqa]

and this is the picture for the component connect to the relay coil

So connection to the relay coil is:
+5v connects to one side of the coil
Source connects to the other side of the coil
0v connects to the Drain

So, in this scenario seems to me that the Nchannel will act as a diode even when gate is low and relay will receive 5v on both coils (S and R) which will avoid it to click ...

Am I missing something, or understanding wrong the diode concept of the MOSFET?

[Kleinstein]

A SOT23 part with 2A code is supposed to be a  SMD version of the 2N3906  (e.g. MMBT3906 or similar from other manufacturer). So it is a normal NPN transistor likely with emitter to ground and collector to the coil.

[eduaqa]

But look at board picture, the PIN number 2 considering left 1, center 2, and right 3 is connected to the ground and the PIN number 3 the one connected to the coil, and pin 1 which would be the base is connected to the signal coming from the logic IC.
But that same as on the MOSFET that I though it was the coil is connected to the right pin, and the ground is connect to the ground

This inversion where the ground is connected to the center Pin is what I still don't understand ... I see some "high side" example of MOSFETs but the Source is always + rail , not coil

[eduaqa]

Multimeter is working now.
So, here some tip that can help others

#1 meter was turning on, but all reading was showing OL in all functions
#2 I open it and I detected that transformer was just generating like 4-5v where it should be generating around 10V
#3 Disconnected the transformer from the board and than it was doing 10.5v, so some short circuit was in somewhere
#4 I detect that one of the 7805 was warming.
#5 I removed it and then the transformer starting generating then 10V as supposed.
#6 Testing the ground and +5v out on the board (with the 7805 removed), it showed 0ohms, so really a short after 7805
#7 here one mistake ... I looked around the 7805 and saw a few components around, so i decide to remove them ... mistake,, the board is multilayer (obvious), but i didn't noticed that on in a first look ... so after removing the components still showing 0ohms
#8 so then I discovered that there was a lot of components, so removing then was not an option
#9 so, with the 7805 removed, used an external power supply and injected 5v on that rail, looking for heating and was easy, those 2 chips I marked initially was heating, removing them remove the 0ohms reading.
#10 however, that reading was still 22ohms, what seemed to be low ...
#11 easy, use the same process with the power supply... not that easy ... this time no obvious heating and the AMP usage on the power supply was not too high ... so what to do
#12 leave it on for some minutes and noticed that the relays mainly those lite orange was warming a little
#13 so I a get a point to focus
#14 measuring the relays saw that all coils was energized at around 5v
#15 another mistake, and here is where you need be patient and understanding before starting soldering/desoldering ... i decide to remove the relays (this was a error, and that was a mistake
#16 ohms went high to 800ohms, now appears good
#17 but, why the relays was energized? I discovered that those sot23 that was connect to 0v and to the coil, was shorting the 0v rail and as the 5v rail was connect direct to the other side of the coil relay it was causing the relay been energized
#18 another big mistake ... looking on the sot23 it had a marking code 2A and a letter E turned 90 degrees ... some google and a found that was a N Channel MOS ... big, big ... big mistake
#19 replaced those sot23 and the relays continued getting 5v on all coils ... so something wrong
#20 here again think before do ... course that would happens N Channel with 0V connect to the gate and relay connect the source and the other end of relay on 5V will act as diode and will energize the coils
#21 so, what now? Thanks to Kleinstein, that called my attention that the device should be a NPN transistor ... actually at the end was PNP, because NPN would not conduct on that direction
#22 remove the N Cnannel and soldered the PNP transistor, the multimeter worked
#23 Also a notice and this is very important the relays that are lite orange are what is called SRCRS relay with 2 coils (Set and Reset), what that means? means that they should not be energized, but, just send a pulse to switch the Set or Reset

Thanks again to Kleinstein.