Hello,

My name is Ty and I'm currently working on developing a test fixture for my product that will allow me to test all of the SSR's on my device.

The DUT is a thermosta that runs off of 24Vac.

My goal for this test fixture PCB is for the test fixture to provide power to the thermostat under test and the thermostat will then turn it's relays off one by one for me to then get a reading if the relay was triggered properly using an RPI pico W.

When the relays on the thermostat close, the thermostat then sends a 24Vac signal to the respective closed relay pin. Given that I am passing the 24Vac signal from the test fixture to the thermostat under test, I have the return line of the 24Vac line available to me.
The way that I had initially intended to get a reading from 24Vac to a 3.3Vdc signal for the RPi, was using bridge rectifiers, a voltage divider and an optoisolated SSR. But I have now tested a batch of initial proto PCBs and am seeing an issue where when one relay closes, then all of the other bridge rectifiers get activated as well. Which is what I don't want. I'm trying to understand if there is any one who can assist in debugging this issue with me?

 

I can't see the big picture. Please add the connections to one of the thermostat .

 

Hey Ron!

Thank you for replying so quickly. I have attached the relay schematic on the thermostat.

Rh, Rc, and ACC+ = 24Vac
C in the screenshot presented in the initial post is the return line for the 24Vac.

W1, W2, Y1, etc. on the thermostat connect to the same net on the test fixture schematic.

 

Is your 24 VAC completely isolated from your DC power?

Where does the 24 VAC come from and what (everything) is it connected to? Does it power the thermostate too (via another bridge rectifier?

 

Hey Michael the 24Vac (should be) is isolated from the DC power.

 

I think when you test one thermostat things work but when you have more than one thermostat there is problems. (?)
I think the way many thermostats are connected together on one AC source is the problem.

 

Michael, to answer your further questions...

The 24Vac comes from an external transformer, that external transformer is run to the pins on the test fixture that then connect to the thermostat. The thermostat is powered via the 24Vac with similar circuitry, bridge rectifier + smoothing caps.

 

Is that a net name at the junction of the resistor divider?

If so, it means that all of your TLP241A's have their anodes connected together, i.e. they'll all get activated at the same time.

 

I see that the opto-isolators are very effectively defeated by all of those ground connections.
There is no reason at all to tie the common of the 24 volt side to the common of the processor side, other than to cause malfunctioning.
The better design would use 24 volt AC relays as loads for the thermostat control outputs, with an indicator such as a 24 volt pilot light as an indicator and spike prevention. Then the contacts can switch the signal to the processor WITH NO CONNECTION to the 24 volt AC side. Quite small relays can be used because the contact current is very low. And in that sort of application the relays will outlast the processor module.

 

In the test fixture there is no reason for the output of the diode bridge have any connection to any power or ground inside
the test fixture -- use the bridge output (floating) to drive the IR diode in the SSR.

 

The benefit of using an actual electromagnetic mechanical relay is that it will be a bit closer to the actual load controlled by most thermostats. Consider that on many gas furnaces the heat thermostat actually controls a solenoid gas valve for heating, and a multi-pole power relay for the compressor and the condensor fan, as well as a power relay for the evaporator blower motor to deliver the cold air..So a bit more load current could reveal a slightly worn out contact that a very low current load would not detect.

 

In the test fixture there is no reason for the output of the diode bridge have any connection to any power or ground inside
the test fixture -- use the bridge output (floating) to drive the IR diode in the SSR.

View attachment 322671

Hey Michael, thank you for your reply. This makes sense. I will test it out. I appreciate your feedback.

 

The benefit of using an actual electromagnetic mechanical relay is that it will be a bit closer to the actual load controlled by most thermostats. Consider that on many gas furnaces the heat thermostat actually controls a solenoid gas valve for heating, and a multi-pole power relay for the compressor and the condensor fan, as well as a power relay for the evaporator blower motor to deliver the cold air..So a bit more load current could reveal a slightly worn out contact that a very low current load would not detect.

Hey Bill, Thank you for your reply. The purpose of this test fixture is not necessarily to get equivalent load to the HVAC contactors that our thermostat typically drives, we do already have a fixture for this matter. The purpose of this fixture, is more for a way for us to test new firmware on our devices and get remote feedback from the fixture that will tell us if the thermostat reacted correctly to temperature changes, etc. I appreciate your feedback and have ordered a couple 24Vac electromagnetic relays to test. Thank you again.

 

While creating the software for those thermostats, it would be valuable to provide a way to remove the "AUTO" mode, where it switches from heating to cooling to hold the same temperature.
Setting the temp to 65 while away during the winter, and then not getting back when it turns warm leads to huge electric bills when the AC tries to hold 65 and the outside gets up to 80 degrees. AND that "AUTO" function will do that. The problem is that fools who don't understand select it instead of the "HEAT" mode.