NI9476 changing output voltage

[hhaamm]

John,

Here are my measurements based on a 8-valve setup. All measurement points are set abt valve #1, when valve # 1 is always included in the "ON group", I have:

# of ON	I_1 (mA)	I_2 (mA)	V_2	V_3
0	0.40	0.10	11.99	0.03
1	29.70	29.70	0.04	11.97
2	55.80	27.60	0.93	11.08
3	46.60	22.90	2.81	9.19
4	62.80	21.00	3.61	8.42
5	52.00	16.70	5.32	6.72
6	40.80	17.00	5.23	6.84
7	36.20	9.70	8.30	3.93
8	28.10	1.10	11.60	0.46

When valve # 1 is always NOT included in the "ON group", ie, with different number of other valves turned on but #1 is always off, I have:

# of ON	I_1 (mA)	I_2 (mA)	V_2	V_3
1	31.80	0.10	11.97	0.03
2	30.80	0.10	11.97	0.04
3	47.80	0.10	11.97	0.04
4	63.30	0.10	11.97	0.04
5	52.77	0.10	11.97	0.04
6	45.71	0.10	11.97	0.04
7	39.80	0.10	11.97	0.04
6	46.07
5	62.10
4	70.60
3	70.60
2	57.10
1	31.20

Note:

1. the current measurement is always not stable. This probably is due to the fly back current from the coil. Especially I_1 in the second scenario, the current is dependent on how I turned on the valves:

either from all ON state to decrease the number of ON valves, or from all off state to increase the ON state valves.

2. The V_1 is always 12.01

3. I have no more than 4 meters of simple copper wire between the output of 9476 to my valves. So I really don't think there is any difference between V_3 and V_4. It's not easy to access V_4 as is, but if you insist on this one, I'll try to get it measured.

3. Please note the value of V_3 (or V_4 if you will). It seems only when 1 valve is turned on, the output level is normal. Although the valve can still be driven by voltage as low as 9 V, but I definitely won't count on the reliability under such voltage.

Please let me know if there is any question of the data is confusing.

Thank you once again!

[John_P1]

Hi hhaamm,

I am still qutie perplexed but I do want to get to the bottom fo this.

When you say the current isn't stable, this is indeed interesting.  I am assuming that you are making your measurements in the steady-state (rather than immediately after a switch).  Do you have an oscilloscope or DAQ card that you can use to measure the voltage across your relays when they are not functioning properly?

Can you try activating each channel individually and confirm that this works as expected?

Do you have a second 9476 to test out?

It does sound like there is something unusual going on that I can't put my finger on.  Can you confirm that you have your circuit connected like the image I posted earlier in this thread?

Best Regards,

[hhaamm]

John,

I was reading using real time reading. I of course can hold the meater reading and get a "stable" reading, but that won't tell the tru story, right? I figure the change is due to the charging/discharging process of the coil, which in turn change the potential of the input end of the valve.

I do have a second 9476 because finally I need to control 64 channels. I haven't done each single test you requested with both of them, but I did try to connect them with and without the pull down resistors at the very beginning and they show the same issue. That's why I tend to believe this is not a defect module, but something wrong abt the design.

It seems this process is going to take very long. Though I'm curious what's going on, I do have a hard deadline for my project. If the problem can't be fixed soon, can you recommend me some other model? As a matter of fact, I was looking at 9477. It was ruled out when I put the orther together for some safety concern. But not if it can work as I want, I'd rather change to that. Please comment.

Thank you!

[hhaamm]

John,

It seems I found the trick. Thanks for your suggestion to double check the wiring and setup: it turns out that the wire I used as common ground, that is connected to the 37 pin connector is broken. Because I connect the ground for each valve separately and didn't use this pin #19 in the connector, I have never noticed this. So I guess for all my previous trials, the ground for 9476 had always been floating!

I'm sorry to take all your guys so much time to scratch your head. But anyway, I guess the problem is solved:)))

Thank you once again!

Best,

Jidong