Relay's not switching using NCD base SW

I’ve been developing software (Java) to control the Fusion 16-Channel DPDT Relay Controller with 16 GPIO or ADC and I2C via the RS232 link.
I’ve had it working, but now it seems to have stopped and no load is on the relays.
I’ve switched to control the unit with the NCD base software via the RS232 link. It reports that the relays are changing state, but can’t hear any of the relays click state. Have also checked with a multimeter to see if they are switching over (they are not).
LEDs on the unit which I believe relate to the state of the relays are on all the time they never switch off, despite the NCD base software indicating that the relays are off
I’m wondering if I’m dealing with a broken unit that was working and trying to figure out how it got broken.
OR
Could I have accidentally messed up the settings of the unit while developing the software? Is there some sort of reset to factory settings in the NCD base software if I have altered something?
Don’t wish to send an item back for repair if it’s a software issue and would prefer to confirm that it’s a hardware issue. Coils measure 360 ohm as per the 5A relay datasheet and have confirmed that the ADC is working as expected.
I’ve used two different bench power supplies (12V) to help confirm that it’s not a power issue, but happy to check things if anyone has any idea on how to solve/confirm this issue.
Thanks

Hi,

It could be a few things.

Software:
Its possible that you somehow turned off automatic relay refresh. Relay refresh determines whether the relays change state. You can find more on this guide: Fusion ProXR Advanced Quick Start Guide - NCD.io

Base Station will have a button to turn off and on automatic refreshing of the relays.

Hardware:
Its possible your power supply isn’t providing enough voltage or amperage. If the power supply doesn’t supply the 12v required you will see that the onboard processor functions correctly, but there isn’t enough power to actually drive the relays.

Its also possible that a power regulator or relay driver on the board have been damaged. This is pretty unlikely unless the board was powered up on a conductive surface such as the ESD bag it ships in, something else was bridging pins on the underside, or a power event happened like a power surge that overcame the protections on the board.

I would check that automatic refreshing is on, the power supply is outputting 12v at least 1 amp, and that nothing could be bridging pins on the underside of the board.

Let me know if any of that resolves your issue.

Thank you so much Jacob for replying.
I tried a few things you mentioned, but for completeness, I redid them.
The PSU that I’ve been using is rated for 2.5A; the bench supply that I’m currently using is 10A.
I measured the voltage using a Fluke 83 III multimeter but it’s an uncalibrated one.
At the input terminals to the board, I measured 12.27VDC with 0.4mV RMS ripple.
The voltage reading between 0V and the coils of the relays measured 11.68V and 11.55V on the other. I assume there is some small leakage current through the low-side drivers down to the 0V around the 360uA mark. I don’t think that’s an issue. As for the 11.68V, a little investigation and assumption show that there is a bridge rectifier (MB2S) being used for reverse polarity protection (diode in series with supply power, I don’t recommend using this type of protection) which has a volt drop around the 0.5V. This explains the slightly lower voltage on the coils from the supply.
I didn’t see any external fly-back diodes across the coils, nor any form of snubber to suppress the back EMF when the coil is switched off. It’s possible that the low-side driver has some protection in it, but it’s not something I would rely on.
On the dual interface fusion logic layer board, there are some connections with silkscreen markings on them(the ribbon cable connection to the lower relay board). Readings from these points are as follows:
3.275V on the pin marked 3.3V
4.98V on the pin marked 5V
12.27V on the pins marked 12V I think there were a few of them.
I didn’t measure the data ones marked SDA and SCL.
The visual checks I did showed no obvious conductive particles, burn marks or damage causing shorts or opens. The board is mounted on 8mm standoffs, has never been directly on a conductive surface and is being used in a controlled ESD environment.
The red status LED is on and the busy LED flashes when the CRD base software is talking to it.
Back to the lower relay board for the next tests.
Pin 1 of the orange resistor pack is 0V, and the other end of the following 8 resistors (22K) reads 9.95V (LEDS are on). The LEDs read 9.95v on one side and 11.66V on the other, giving about 1.7V across them. Sounds about right to me for an LED. The 11.66V is due to the diode drop with the reverse diode protection (MB2S) mentioned earlier.
The TPIC6B595 looks to be the low side relay driver chip checked pin 2 which is the supply pin to it and reads 4.99V. Pins 10,11 and 19 were checked to read 0V and not floating.
The ribbon cable between the lower and upper PCBs was checked removed, inspected and reseated to ensure that it was not loose.
Using the software PROXR advanced Relay control command set with an RS232 link. I did also try the ethernet link as well.
Verified that the refreshing was on.
Turned the auto-refresh off and on again and checked the control of relays, still not working.
Tried with manual refresh, no improvement.
Measured the voltage on the TPIC pins 4 to 7 as 11.52V did not change when the software was indicating that the relays were on or off.
checked the relay status button to confirm if the relays are on/off. there was no change in the physical relays or the LEDs.
Used the relay flasher function at different speeds, no change was seen on the relays, LEDs or the drain outputs of the TPIC chip.
Tried using the Fusion series override controller for banks 1 and 2 to change the relays and had no success in changing the state of the relays or LEDs.
I’ve used a thermal camera to see if there are any components overheating and nothing showed up as being stressed.
By following PCB tracks buzzing out connections it was possible to figure out the connections from the TPIC to the microcontroller.
There is a diode pack on the control lines, I assume for ESD reasons and they seemed to be functioning as I would expect.
Using an oscilloscope I checked the control lines to the TPIC low-side driver chip. I set the software to flash mode the relays, so it would be constantly sending messages out.
The SRclk line seemed to be sending out 7 pulses and then remained high. The datasheet shows this should be 8 pulses and held low.
The serial data line appears to be good and synced up with clk pulses. I didn’t spend the time to decode the data to confirm if it was exactly as I would expect.It did remain high after the data was sent and the datasheet shows that it should return low, but I don’t see this to be an issue with what I’m experiencing.
The Rclk line looks like it’s an inversion of what is shown in the datasheet for the TPIC chip. The datasheet shows a pulse going high and then low. When measured it showed a constant high with a low pulse.
The not EN (not G) line was constantly low, which is what I would expect looking at the datasheet. it was high when the unit was first powered on but dropped low after a period.
Ideas and other things to try out?
Thanks

Hmm, what all functionalities are you using on the Fusion board (Taralist, Reactor)? Does your fusion board have a TLEE module? Were there changes made to the baud rate?

You can see if changing the Midnight backup buffer options helps recover activation. Go into the Fusion Master Control Panel in our Base Station Software (Start - NCD.io 2) and make sure that Taralist is currently on and is set to be on at powerup. Then go into the Taralist Enhanced Event Controller in Base Station. Then click on the MORE button in the top right hand side. Click on Advanced Interface. Click on Midnight Backup Buffer. Make sure that all boxes are unchecked on the Midnight Backup Buff panel and then hit save.

Hi Jacob
I’m not using a TLEE module, not using the TARAlist or Reactor. Just the Fusion XRPro commands. We are running the unit at a baud rate of 19200 and have tried 9600 as well as 115200. Will try 57.6K later today. I’ve tried the restore factory setting under the Reactor menu, but this had no effect. I did check the TARAList is set to on at startup. As we don’t have the TLEE module the NCD base software will not allow access to the TARAlist enhanced event controller.
I’m a little torn between a firmware issue on the PIC micro that’s on the fusion control board which controls lines to the TPIC relay controller chip on the lower relay board or that the TPIC controller chip is damaged.
Thanks