TIC10024-Q1: TIC10024-Q1 cable length - reliability, emi etc...

Hello Jan,

I want to clarify a point to make sure I fully understand your question and application. 

I understand that the package distribution box is and generally how it works from your description.  This point is clear and there will be a signal applied to the solenoid to open a particular box once a PIN is entered by an individual.

However, when you say you want "to control state of switches in package distribution box" do you mean that you want to use the TIC10024-Q1 as the source for the signal sent to the solenoid to open the switch?  Or do you want to use the TIC10024-Q1 as a monitoring device to know when the switch has been opened or closed?

The TIC10024-Q1 is designed to be a detection device to indicate the open/closed state of switches and not as a device to change the state of switches. It will source or sink a small amount of current to the switch and measure the voltage to determine whether the switch is open or closed.  It does this for each input pin or switch one at a time.  If you want to monitor the status of the box switches, then this device could do that.

However, if you want to control the state of the switches in the box and open a particular switch after a PIN is entered, then this is not a good device for this application and you would need a device that could source or sink enough current to control the solenoid.  How much current would the solenoid need?  There are a variety of multi-channel High Side Switches, Relay Drivers, and constant current LED drivers that could all be options in this application.

Regards,

Jonathan

Hello Jonathan,

thank you for your answer. I see that my question was little bit confusing so I will try to explain it better.

Let me share a simplified diagram of the system:

I would like to use TIC10024-Q1 to monitor state of the status contacts - they just provide a feedback that the lock was succesfully unlocked by the solenoid. I will use low side switches to control solenoids.

I have two questions:
1) is the distance between DOOR lock and TIC10024-Q1 okay? My plan is to use recommended capacitor and resistor to protect TIC10024 input against ESD, but I am not sure if I should use something else because of the wire length (2x7m).

2) Is the contact configuration of statust contacts (normaly closed - when the DOOR lock is locked status contact is closed) okay for TIC10024? Isn't there a risk of overheating device when it runs in continuous mode (it probably depends on the current, right?) ? I can use fan or heatsink if needed.

Thank you,

Jan

Edit: Please note the Q1 low side switches will be part of the same PCB like TIC10024 - i just made a mistake in diagram..

Hi Jan,

Thank you for the clarification and I apologize for my initial confusion to your initial question.

1) The distance of the cable should be ok.  The resistance and capacitance of the actual wire used will contribute to the impedance of the switch itself.  However, there are several current options available and several voltage thresholds that can be set which should accommodate the 7-10m.  Do you know what guage and type of wire you will be using?  Do you also know what the contact resistance or switch type you will be using? If so I can try to setup a quick test to verify your conditions.

2) If you have a high current setting, there is a risk of overheating the device.  However, a constant current mode is not required and the Polling Mode is recommended with the high current settings.  The device will only source/sink the current on the pin when it is making the measurement.  This way the overall power is reduced and unless there is a large RC time constant on that pin that would prevent an accurate measurement, there should be no difference to the measurement accuracy.

The optimal settings for the current and voltage thresholds will depend on the system properties such as the wire and switch impedance, etc.  Once those are known, a more accurate assessment can be made.

Regards,

Jonathan

Hello Jonathan,

it is okay - my description wasn't very accurate :)

Sadly our customer decided to use some un-name chinese locks so there is no information about switch impedance but it is some kind of chinese snap action switch with lever:

It has some markings but I wasn't able to find any datasheet or product information.

Our plan is to use unshielded 2 core cable because of price - something like this: https://cz.farnell.com/lapp-kabel/0028402/cable-liyy-2core-0-34mm-50m/dp/1204323?gclid=Cj0KCQiAzsz-BRCCARIsANotFgOjDuGGCgylW4nMdgqVBkHYN197bZLcQqgGC1rh5ZaHJV9a6k1gGRsaAlEAEALw_wcB&mckv=shWy2iVjY_dc|pcrid|201179875148|kword|0028402|match|p|plid||slid||product||pgrid|48509977848|ptaid|aud-899780980989:kwd-316772251899|&CMP=KNC-GCZ-GEN-SKU-MDC

As we don't have the switch type it doesn't make sense to do any testing on your side, right? So any other recommendation except 7-10mA current and polling mode?

Thank you,

Jan

Hi Jan,

Thanks for additional information.  I have some similar lever type contact switches that I can use for a test that should be good enough to get an idea of the current/voltage settings that may be needed.  But I think the 7-10mA and polling mode should be adequate for this application.  I will get a quick test put together to confirm and I will follow up early next week. 

Regards,

Jonathan

Hi Jonathan,

thank you very much for this. Would be possible to send me a quick hand-drawn schematic and photo of your setup after you test it (if you won't use the recommended schematic from datasheet)?

Thank you in advance!

Jan

Hi Jan,

Yes, I will provide you all the information about the test setup.

Regards,

Jonathan

Hello Jonathan,

did you have time to test it?

Thank you,

Jan

Hi Jan,

Yes, I was able to get a quick test, and there should be no problem with 10m of wire and this type of switch but I have not tested with the series resistor or capacitor for ESD protection which I had planned on doing before I responded to you.  However, what I have done is the following:

I used a single twisted pair from an Ethernet cable that was 10m in length which is 26AWG.  The cable you suggested was 22AWG so it is a bit thicker and should have lower resistance than my test.

I soldered a switch across one end of the wire and connected the other end of the cable to the Input 0 (IN0) channel of the TIC12400/TIC10024 EVM.  I configured it with the GUI to source the lowest current level (0mA) and set a 2V comparator threshold.  I was able to detect the switch contact every time I closed the switch which made a connection with GND. 

The EVM has 0-ohm resistors populated for the Resd resistor, and the Cesd capacitor is not installed.  I plan on adding a 50 ohm resistor and 0.01uF cap tomorrow just to see if there are any changes to the current or voltage level thresholds.  But since we are able to use the lowest settings at the moment, I am certain it would work in the system and there is plenty of margin to make adjustments.

I have never used a cable this long between an switch and this device and I am curious to see if there are any changes to the levels which I will test tomorrow.  But at the moment the answer to your question is that it will work in the system.

Regards,

Jonathan

Hello Jonathan,

thank you for your testing! 

Isn't it better to use wetting current higher than 0mA? The idea to use wetting current at least 1 mA is to have better imunity against RF and other noise.

Also my switches are normally closed, but I guess that your test proved it works, but I will have inverted logic.

Thank you,

Jan

Hi Jan,

Yes, you are correct that a higher current would give better immunity against RF and other noise and you should do so in your final application.  However, you have asked me about the feasibility of using the TIC10024-Q1 with a 10m cable and therefore I need to look at the extreme cases where it may not work in terms of wetting current and the comparator voltage threshold. 

I have tested with a 220 ohm series resistor and a 0.01uF capacitor to mimic some ESD protection components you may use and I have found that I can still detect the switch with all wetting current and all voltage threshold setting combinations.  Therefore you should be able to easily find a setting that works well for you in your final application.

I suspect you will have a large coupled noise spike onto the switch line when the solenoid switches, and therefore you would want to use a larger current and the Cesd capacitor to help prevent a false detection of the switch if the device polls that switch input during the moment of the noise.  However, there are also internal settings that require multiple consecutive polls to indicate a valid switch change that can be used as well.

I am including a picture of the test setup for your reference. 

Regards,

Jonathan

Hello Jonathan,

thank you again for a big help!

I also suspect coupled spikes when the solenoid switches (solenoid and status switch wires are in one cable) but I don't think they will cause any problem as the normal operation (locked state) means that current flows and when it is unlocked the current doesn't flow. Also the operation of the device is soooooo slow - I will make a pulse on solenoid wires to open the lock and then I can wait like 200ms to control the state of the switch so I am not very worried about the spike problem.



Will you make more tests to test extreme cases or you're happy with the results you have? Also Is TIC10024 good for our application or is it better to use TIC12400 for any reason?

Thank you,

Jan

Hi Jan,

I don't plan on making any additional measurements, and it appears that with a similar type of switch, wire, and length, that there is plenty of margin to get a system configured properly.

The only big difference between the TIC12400-Q1 and the TIC10024-Q1 is that the TIC12400-Q1 has an additional ADC option that can be used instead of the comparator option.  This is useful for multi-position resistor coded switches because it can detect the switch position as a function of the different voltage drops across the different resistors.

In your application it sounds like an open/closed style of switch would work best with the comparator and therefore the TIC10024-Q1 would suffice.

Regards,

Jonathan