Simple circuit with Photointerrupter has got me baffled...

Thread Starter

pureanalog

Joined Oct 8, 2010
36
In post#5 I specifically mentioned in the drawing that "Only two wires, i.e. the +5V and GND" should be connected.

Can you confirm you have followed this requirement in your checking?
Wait a minute!

When you say connected, you mean connected to the ribbon cable?

If so, no. The ribbon cable was attached and connected to all pins!!

All were connected.

If this is what you meant, let me know and I will try again, when I go back to the instrument in an hour or so and get back with results.


I will try, getting a mirror!


Thanks so much for helping me troubleshooting it.

I so very much appreciate it!!! :D
 

eblc1388

Joined Nov 28, 2008
1,542
When you say connected, you mean connected to the ribbon cable?
Yes.

If so, no. The ribbon cable was attached and connected to all pins!!
The idea was to remove the possibilities of the "upstream" circuit affecting your measurement of the 4 photo-interrupters.

If this is what you meant, let me know and I will try again, when I go back to the instrument in an hour or so and get back with results.
Please do.

I have a feeling that you will soon discover you are trouble shooting a perfectly good board, where the problem lies elsewhere.
 

Thread Starter

pureanalog

Joined Oct 8, 2010
36
See post #3. ;)
You guys were absolutely right unfortunately, and I say unfortunately because the problem lies somewhere else before the little pcb board!!

I measure 5Volts and 0Volts when I interrupt....

The ribbon cable comes from a motherboard on the instrument. Any clues on what I should be looking for there on the motherboad?

Now what could be causing such a behavior?
 

retched

Joined Dec 5, 2009
5,207
Any mechanical devices should be checked first.

Check the components that attach through the ribbon cable to the part where the 5v should not have been before the interrupt.

Then power off, check for continuity in places where there should be none.

This will help you track a component that failed short.
 

Thread Starter

pureanalog

Joined Oct 8, 2010
36
http://img189.imageshack.us/i/moto0210.jpg/

http://img524.imageshack.us/f/moto0209.jpg/

Here above are the pictures of the intermediate pcb between the pcb I was working on and the motherboard. It hosts another two sensors with a circuit identical to the pcb we were discussing. Those two seem to work well too.

Then the wide ribbon cable connects to the mainboard.

I traced back the 5V and Sensor points that gave me 0,7volts when all was connected, and removed the main ribbon cable from the motherboard and measured those two to find 0 Volts.
 

retched

Joined Dec 5, 2009
5,207
Second picture moto0202.jpg:

Is that a tear in the trace on the bottom-right edge by the screw hole??

Looks a bit weird. Also, check those optos the same way you checked the others.

And if you can update or re-install the firmware, I would do that. Others may say to wait, but If it is a firmware problem, that should solve it.
 

Thread Starter

pureanalog

Joined Oct 8, 2010
36
Second picture moto0202.jpg:

Is that a tear in the trace on the bottom-right edge by the screw hole??

Looks a bit weird. Also, check those optos the same way you checked the others.

And if you can update or re-install the firmware, I would do that. Others may say to wait, but If it is a firmware problem, that should solve it.
No it is ok. Its just the photo. I'll try the re-installing of the firmware tommorow.
 

GetDeviceInfo

Joined Jun 7, 2009
2,194
For point C I get 0,7 Volts when nothing is inbetween the IR sensor. When I block it I get 0 Volts
I measure 5Volts and 0Volts when I interrupt....
With everything connected, and measuring the indicated point C, what voltage swing are you seeing between actuated and non?

you are sinking a sourced voltage. You need to determine if your sourced, then determine that your sinking. you say you manually sunk, at what point? If you get full swing at the sub PCB, I'd meter the signal line at it's connector on the motherboard and look for full swing.
 

Thread Starter

pureanalog

Joined Oct 8, 2010
36
With everything connected, and measuring the indicated point C, what voltage swing are you seeing between actuated and non?

you are sinking a sourced voltage. You need to determine if your sourced, then determine that your sinking. you say you manually sunk, at what point? If you get full swing at the sub PCB, I'd meter the signal line at it's connector on the motherboard and look for full swing.
Just did the measurements again:

With everything connected at the indicated point C, the voltage swing is between 0 and 0.7 Volts for blocked and unblocked respectively. For all other points it is 0 and 5 Volts.

When only the 2 points (5V and GND) are connected I get a voltage swing of 0 and 4,85 for the point C and 0 and 4,92 for all others.

When only the 2 points (5V and GND) are connected the voltage across the resistors is 3.66 Volts for point C and 3.73 Volts for all the others.





*I updated the firmware but it didn't help.
 

Thread Starter

pureanalog

Joined Oct 8, 2010
36
you are sinking a sourced voltage. You need to determine if your sourced, then determine that your sinking. you say you manually sunk, at what point? If you get full swing at the sub PCB, I'd meter the signal line at it's connector on the motherboard and look for full swing.
Can you determine the signal line. What would the point of measurement be? And with everything connected?
 

retched

Joined Dec 5, 2009
5,207
Maybe its time you tell us the malfunction symptoms.

You skipped telling us in the first post, but that my lead to investigating the proper part.
 

Thread Starter

pureanalog

Joined Oct 8, 2010
36
Maybe its time you tell us the malfunction symptoms.

You skipped telling us in the first post, but that my lead to investigating the proper part.

Alright. Here it goes.

The instrument is an autosampler for liquid samples. It has a tray with 100 positions in which you place the vials with the samples.

There is a moving arm that moves in 3 dimensions, X, Y and Z. Left and right, and up and down when the arm with the syringe dips into the vial.

So there is a piece of metal mechanism that when there is a vial in the tray that the mechanism grasps on and it is pushed back and blocks the so called Z sensor (the one malfunctioning).

The Z sensor gives the instrument a signal that a vial is present or not depending on if it was blocked or not. So the problem is that the instrument constantly gives a signal that there is no vial in any position. This way the instrument is useless...

There are test points on the pcbs that are marked as "GND, Z sensor, Z home, Y Sense, X Sense etc" . I measure 4.20 and 5 Volts when I interrupt and let go at the problematic sensor. So it seems to respond but not properly.
 

Thread Starter

pureanalog

Joined Oct 8, 2010
36
So the z sensor is working.. The motor or servo that moves the syringe up/down is not?
No, I confused you here.

The motor works. And as the syringe dips, the mechanism that should block the sensor to let it know that there is a vial present blocks it. But the instrument does not take the signal apparently.

Also there is a diagnostics menu on the instrument where you can monitor a few of those sensors live. You get a 0 and a 1. So when you put a piece of paper in the sensors you can see that the other sensors respond, but the Z sensor does not...
 

retched

Joined Dec 5, 2009
5,207
ahh. I see.
Can we get a make and model number of this unit?

We may be able to get some schematics for it somewhere on-line. That would help more than guessing around.
 

Thread Starter

pureanalog

Joined Oct 8, 2010
36
ahh. I see.
Can we get a make and model number of this unit?

We may be able to get some schematics for it somewhere on-line. That would help more than guessing around.
The model is a Dionex AS50 autosampler. I doubt though that you can find schematics. I was talking with a guy that used to work for Dionex, and he told me that the company does not give schematics to their own service representatives!!! They just replace boards. More cost effective and profitable for them apparently... Unfortunately the boards are too expensive for me to afford.
 
Top