NI9476 changing output voltage

[hhaamm]

Hello all,

I've been fighting this problem for a while and no one seems to know an easy fix, even the application engineer I talked to.

I have a NI9476 module and I plan to use it to drive 32 solenoid valves. The voltage required by each of the valves is 12V DC, current consumption ~ 30 mA.  I connect a 12V 5A DC power supply to the V_sup of my NI9476, and a 10K pull down resistor between channel and ground, hoping I can control all 32 channels in parallel. However, this type of setup only works for several channels in paralles (<4). Once I have more channels turned on, the voltage across each channel drops significantly. For example, once I have 8 channels turned on, the voltage for each channel is only ~ 7V!

One NI application engineer told me output voltage is supposed to be dependent on the number of channels turned on and the load for each channel. I don't know abt that and feel it's kinda weird if it is so. Could someone with experience with this product comments on this? Thanks a lot!

[John_P1]

Hi hhaamm,

Do you have a link to a schematic of the 32 solenoid valves?  Of specific importance:

    Do they have built-in flyback protection (e.g. Schottky diode)?

    What about a built-in method to limit the current in the steady-state?  A 50 Ohm (or greater) resistor in series would probably be acceptable given the 250 mA per channel spec of the 9476.

If not, I suggest adding these things yourself to the circuit.  If the 30 mA is what you are measuring per channel and there is no current-limiting resistor, then the module is probably in its overcurrent state and 30 mA is what it is passing through.

The 10 kOhm resistor from DO to COM (which is how I read your statement) doesn't really do much--I'm assuming you installed it based on this Developer Zone article which doesn't really apply for driving inductive loads.

Best Regards,

[John_P1]

Addendum to previous post:

A high enough coil resistance could be sufficient as well.  Do you know what that is for your solenoid?  A link to some information about your solenoids along with a schematic of your connection would be helpful to ensure we are on the same page.

Best Regards,

[hhaamm]

John,

Thank you for your comments.

The model of my valves is VQZ115-6L01-CP, made by SMC. In the attached diagram, it is shown that fly-back protection diode is included. I don't have the input resistance for the diode though. By simply measuring it by a ohm meter, it's 389 ohm.

I tried to remove the pull-down resistor and drive the valves w/9476 and 12V DC only. It's interesting that even if I only need to test one valve, I still have to connect multiple valves, say 8, to the output. Otherwise, it won't work at all. With that, the current I measured at the from my power supply is 32 mA when one valve is turned on. The voltage drops when I turn on multiple valves, so that it is still impossible to keep all 8 valves on. I don't know if this can be fixed if i connect all 32 valves. The current  changes from 32 to 50, 66 and 59 mA when I have 2, 3 and 4 valves on at a time. I can't have more than 4 valves on though.It really feels the current is limited somehow. But since 250 mA is the limit for each channel, why you say 30 mA makes the module in a overcurrent state?

I'll try the current limit resistor later today.

Thank you once again!

[hhaamm]

---

@John P wrote:

Hi hhaamm,

Do you have a link to a schematic of the 32 solenoid valves?  Of specific importance:

    Do they have built-in flyback protection (e.g. Schottky diode)?

    What about a built-in method to limit the current in the steady-state?  A 50 Ohm (or greater) resistor in series would probably be acceptable given the 250 mA per channel spec of the 9476.

If not, I suggest adding these things yourself to the circuit.  If the 30 mA is what you are measuring per channel and there is no current-limiting resistor, then the module is probably in its overcurrent state and 30 mA is what it is passing through.

The 10 kOhm resistor from DO to COM (which is how I read your statement) doesn't really do much--I'm assuming you installed it based on this Developer Zone article which doesn't really apply for driving inductive loads.

Best Regards,

---

John,

I've added a 50 ohm resistor in serieal to each channel and it seems that still won't fix the problem.

I have connected 8 valves, each w/ a 50 ohm resistor between its ground and the common ground. I have the same 12 V DC power supply. with those, I can turn on no more than 5 valves at a time. And the current are measured as follows:

1 valve on: 30 mA

2 valves on: 56 mA

3 valves on: 66 mA

4 valves on: 83 mA

5 valves on: 72 mA

6 valves on: 62 mA (at this time, only the leds on valve 1-4 are lit, which shows an on state. while I have turned on 1-6 from MAX).

Again, it seems something is limiting the current. Any comments?

[John_P1]

Hi hhaamm,

It looks like I jumped to some conclusions prematurely.  The coil resistance is evidently high enough to not require an external resistor (although... do you have a link to a data sheet to confirm that it should be ~380-390 Ohms?). 

So, that gives us the following:

       

Could you give us some more information by answering the following questions on the above diagram:

    1.  What is the model of power supply that you are using?

    2.  Measure the current and voltage at the above test points (V3 measured at the 9476 connector and V4 measured at the solenoid connector) for the single relay case.

    3.  Measure the same as above for multiple relays (say, 8).

Looking forward to your response, this one does have me scratching my head a bit as well but hopefully the above information can help us narrow down what is happening.

Best Regards,

[hhaamm]

John,

Thank you so much for your effort. I'll do the measurement later today. But let me answer the easy one first:

the power supply I'm using is made by Carlo Gavazzi, mode # SPD1260.

Hopefully I'll get back to you with the measurement ersults later today.

I do not have any document or data sheet that shows the input resistance of the valve. The ~ 390 ohm value is simply measured by an ohm meter. I did measure multiple valves though and the readings are same from valve to valve.

[hhaamm]

John,

Per your request, here are the results:

1 valve connected, turn on and off:

Measurement     On                   Off

I_1 (mA)              1                      1

I_2 (mA)              1                      1

V_1 (V)              12.01              12.01

V_2 (V)              11.91              11.97

V_3 (V)               **                   0.042

8 valves connected, turn 1 valve on/off only:

Measurement     On                   Off

I_1 (mA)             34                      1

I_2 (mA)             33                      1

V_1 (V)              12.01              12.01

V_2 (V)              0.034              11.99

V_3 (V)              11.98               0.023

**: value changes from 0.068 to 0.104 slowly (over several minutes) and gets stablized around 0.104, w/ +/- 0.001 fluctuation.

Abt V_3 and V_4, coz I simply have a wire connecting the test point of V_3 and V_4, I really dont see why there should be any difference.

---

@John P wrote:

Hi hhaamm,

It looks like I jumped to some conclusions prematurely.  The coil resistance is evidently high enough to not require an external resistor (although... do you have a link to a data sheet to confirm that it should be ~380-390 Ohms?). 

So, that gives us the following:

       

Could you give us some more information by answering the following questions on the above diagram:

    1.  What is the model of power supply that you are using?

    2.  Measure the current and voltage at the above test points (V3 measured at the 9476 connector and V4 measured at the solenoid connector) for the single relay case.

    3.  Measure the same as above for multiple relays (say, 8).

Looking forward to your response, this one does have me scratching my head a bit as well but hopefully the above information can help us narrow down what is happening.

Best Regards,

---

    

[John_P1]

Hi hhaamm,

I mentioned V3 and V4 separately since I wasn't sure if you might have long wire runs etc.  The values should indeed be almost exactly the same.

In the 1 valve case, do you mean that the solenoid is only drawing 1 mA when it is being driven or is this a typo?

In the 8 valve case, can you try when driving multiple valves so that they start to exhibit problematic behavior (not turning on).  The numbers you have given are the expected behavior and don't show any problem--I'm more interested in what happens in the problematic case.

Best Regards,

[hhaamm]

It's not a typo. With only one valves connected, it CAN NOT BE TURNED ON and the current is always 1 mA. I definitely this is something not normal, not sure if it's correlated with my initial problem though.

I'll get the measurement for multiple valves later today.

Thank you!