Thermocouple amplifier tests fail


 
M

Mirko Petrovic

New member
I have only begun assembling the board and I immediately encountered a problem. I tried to extrapolate the instructions for 4.2 onto 4.3 for thermocouple testing after I assembled that part. The first test with ohmmeter passes - all joints seem to have been soldered properly. Then, as instructed, I applied 9V but the output measured only 4.3V. The third test also failed - when I shorted the inputs, the output was almost unchanged at 4.1V.

I should also mentioned here another issue but I do not think it is relevant. This being my first SMD soldering, I managed to rip off the pin 4 from the amplifier! But both in the schematics diagram and chip specification the pin is "not connected" so it should not present a problem.

I can try to get a picture of the board if someone thinks it may help, although I only have the iPhone and not a real camera so I am not sure if anything legible will come out with such small components.

Any input will be highly appreciated.
Thanks
 
You need a good magnifier to get a close look at the amp. What I do after I get one lined up correctly is to add alot of solder on the opposite side I used to tack it down. Then I add solder to the other side. Then you need solder wickets works best to also have a bottle of rosin to add, but it's not needed. Use the wick to remove all the extra solder and take a very good look at it. You shouldn't see any solder inbetween the legs, except for 2, 3 and 7,8. You are correct in that 4 is not used and that the leg I usually use to track the amp to the board.
 
First, make sure you're probing the proper points for the 4.3 board. The easiest way to do this is to use the probe expansion header J2. +9V to the 3.3V pad, GND to GND, and 0 is your test point. Note that this assumes you do not have any of the other through-hole components on the board, just the thermocouple parts. You can also verify you have 9V at pin 7 on the amplifier and GND at pins 2 and 3. If that checks out and you're still getting 4.3V at the output, then the only thing you can do is to disconnect power and follow the traces to make sure they have continuity as expected. Everything checks out there still? Seems like the thermocouple amplifier is damaged.
 
Thanks for the quick reply.

John, I have inspected the soldering visually the best I could and everything appears OK. I was able to verify that 2, 3 and 7, 8 are still shorted, but not others.

Yes Bryan, there are no through-hole components, only the surface mount capacitors, resistors and the amplifier, 8 in total. And these are the exact points I used in testing last night. I have repeated them now with the same result except I was typing the voltage yesterday from memory and this time I wrote it down. The actual readings I am measuring are 4.12V and 4.06V.

And I have also verified using the ohmmeter with the power disconnected that all pins from the amplifier are connected to all points on the board indicated from the Eagle program using its "Show" function.

The only thing I can do now is order another chip which I hate because the shipping will cost me as much as the component itself! Unless someone knows a place in Portland where I can by them directly...
 
Since you mention ripping off Pin 4 from the amp, ruled out shorts and verified continuity, I would guess the TC amp is dead... which is a bummer because, as you state, shipping will kill ya for that tiny part....
but if you shoot a pic with your Iphone zoomed in on the TC circuit maybe something will jump out that is an issue. Otherwise, be careful removing the bad IC, use a magnifier to see what you are doing. You need to remove the bad amp cleanly, without removing any of the traces with it, and solder the new chip in without any shorts...
 
Apply lots of flux to both sets of pins and you should be able to heat it up and float it off the pads with ease.
 
I have already desoldered the chip, but I did take a picture in case someone asked (glad you did!). I have also got one after the surgery. As you can barely see due to the camera quality, I totally killed the pin - even the pad was gone with it!

https://drive.google.com/open?id=0B7E13k7gVKiDWXVzOHpGREtMc0E
https://drive.google.com/open?id=0B7E13k7gVKiDRGJxRWZxR2dqYk0

At least I was able to find a retailer that offered a lower price than the other major players and on the top of that with free shipping - stock up, boys and girls!

https://www.arrow.com/en/products/ad8495armz-r7/analog-devices
 
I have actually bought them from Verical (who owns Arrow) for under $3 Q100 before, and the last batch I ordered was from Mouser was just $3.50 for 100. It is really strange how the price on these changes so radically. It isn't like this is a brand new product or the company is going out of business!

Hope the new amp fixes your problems!
 
$5 bucks shipped isn't too bad to recover from a situation like this. Looking at the pics the 1n cap to the right of the amp looks like the soldering and placement is a bit hokey, you might want to touch that one up a bit.
 
It appears I am doing something fundamentally wrong here. I have installed the new amplifier (still nothing on the board beyond its supporting resistors and capacitors) and I measure 7.35V on the output pin while feeding it with 9.42V. The second part of the test seems to work though. When I short the thermocouple inputs, the output does drop to the "expected" 7.29V.

Does this make any sense?
 
You must log in or register to reply here.

 

Back
Top