[Stuff-Builder]

I just set up a CNC4PC solid state relay board. Basically I wired the board into two duplex receptacles, added the 5v power supply, blah blah blah. So the board seems to work fine - turns the appropriate receptacle on and off when the keys are pressed in Mach.

My question/issue is this: the input signal to the board is an active low. The problem is that when you disconnect the board from the parallel port - or turn the computer off - the relays then all close. I found this out this morning when I shut the CNC computer down and the spindle on my mill promptly jumped to speed. Not exactly the most soothing occurance.

Did I do something wrong in my wiring? Is this normal for an active-low output? Anyone used this board and faced this problem?

Thanks,
Scott

[mgamber]

I just set up a CNC4PC solid state relay board. Basically I wired the board into two duplex receptacles, added the 5v power supply, blah blah blah. So the board seems to work fine - turns the appropriate receptacle on and off when the keys are pressed in Mach.

My question/issue is this: the input signal to the board is an active low. The problem is that when you disconnect the board from the parallel port - or turn the computer off - the relays then all close. I found this out this morning when I shut the CNC computer down and the spindle on my mill promptly jumped to speed. Not exactly the most soothing occurance.

Did I do something wrong in my wiring? Is this normal for an active-low output? Anyone used this board and faced this problem?

Thanks,
Scott

That's not good. Do you have it wired to NC or NO?

[Stuff-Builder]

Well there wasn't really any option for normally open or normally closed. Just two terminals for each switch.

The manual is at http://www.cnc4pc.com/Solid_State_Relay_Board.htm, then pick the pdf manual. No mention is made of this in the manual. According to the manual

It seems to have to do with the inputs being active-low, but it does essentially make the board normally-closed.

Ideas?

Scott

[mgamber]

Well there wasn't really any option for normally open or normally closed. Just two terminals for each switch.

The manual is at http://www.cnc4pc.com/Solid_State_Relay_Board.htm, then pick the pdf manual. No mention is made of this in the manual. According to the manual

It seems to have to do with the inputs being active-low, but it does essentially make the board normally-closed.

Ideas?

Scott

Without product bashing and based on only the information you have given, I'd have to say get a new board. That is totally unsafe.

[bcromwell]

I have the same board, and I am 99.99% sure mine does not work this way. Make sure you have the ground going to both the enclosure and the rs232 gnd pins, if you don't the gound might float after removing the cable turning off the pc. I will test mine tonight but I have rebooted and turned of my computer a lot and never had the spindle or vacume come on.

[bcromwell]

I have the same board, and I am 99.99% sure mine does not work this way. Make sure you have the ground going to both the enclosure and the rs232 gnd pins, if you don't the gound might float after removing the cable turning off the pc. I will test mine tonight but I have rebooted and turned of my computer a lot and never had the spindle or vacume come on.

Just found in the Product manual,

Wiring:
This card works as a set of 4 normally open switches that enables the connection if the power is supplied to the card, and a signal is provided to the appropriate input pin. Please make a note that this card works only for AC devices.

You might have miswired something...

[Stuff-Builder]

I got some info from Arturo - the guy who built the card. He recommended using a safety charge pump to switch the 5v off when Mach is not running.

He says the parallel port can do wacky things when Mach isn't running.

I might just go ahead and ge the charge pump card. Alternatively I already have a switch wired to turn my stepper drivers off (I get tired of the high-feq. whine). Maybe I'll just wire the 5v relay supply to this switch as well. I never leave the stepper drivers running when I'm not using Mach - too noisy. That's probably the 5 cent solution.

Scott

[PoppaBear10]

I have this card as well, mine doesn't do that. It may indeed be your wiring. You can also do a "Cheap" enable circuit, just have a TTL relay switch your power supply to keep your relays Charged when the enable signal is not present. Mouser makes some solid state relays DC/DC that will work with a 5v signal and then has the power side upto 5-30Amps.

Scott

[Stuff-Builder]

Something is very wrong with it right now. I just did some testing and I was getting power out of a relay that isn't even hooked up.

I think I may have done something wrong with grounds (???), I don't know. Arturo has been good about getting back to me and asking more questions. He says he hasn't seen this problem before and that the board shouldn't behave like this (a good thing!).

Scott

[Stuff-Builder]

I've been discussing the problem more with Arturo Duncan at cnc4pc, and he (and I) decided the board must have been damaged in shipping. It's just not behaving the way it should at all.

I give a big thumbs up to Arturo for replacing the board with a new one and picking up the shipping to boot. I appreciate the good customer service!

I did go ahead and buy a safety charge pump board from Arturo as well.

So, hopefully that will take care of everything.

Scott

[Al_The_Man]

Something is very wrong with it right now. I just did some testing and I was getting power out of a relay that isn't even hooked up.

Keep in mind that a solid state relay will measure a voltage out with no load connected to it when using a high impedance meter like a modern digital voltmeter, this is due to the inherent leakage of the SSR and the high impedance of the meter. It's is just that this leakage will not support any kind of normal load so you do not see it normally.
Al.

[Stuff-Builder]

Al,

Can you explain better? Use small words for a spark-clueless mechanical engineer! Talk to me like I'm four.

BTW, my multi-meter is an old Radio Shack analog unit bought by my dad sometime in the 70's.

With the solid state relays, if I plug just a simple test light into the outlet wired to the relay would it come on from this leakage? Is that a small enough load?

Mass confusion...

Thanks, Scott

[Al_The_Man]

Scott, If you have a moving coil meter, it can still register voltage from opto output to neutral if it is sensitive enough, I have a Grayhill AC Opto output module here that states Max. off-state leakage 2ma on 120vac. Normal load from 30ma to 3 amps.
So it means that if you hooked up a load that resulted in only 2ma (2watt lamp) it would conduct to the load regardless of wether the opto was on or not. It may not be the situation in your case, but it is something to be aware of when measuring voltages where solid state devices are concerned.
Al.

[Stuff-Builder]

Al, thanks for the reply.

The leakage probably could have lit my outlet tester. Do you think the bulbs are 2 watt? I'm guessing I would be better to use my 120v/60 watt drop light as my "test load".

I did de-bug my parallel port setup with a couple of LEDs and resistors, and the signals from the computer are behaving exactly the way they should. So that's good.

I found out that my 120 ac plug had the polarity reversed, but that shouldn't be a big deal. Beyond that I guess I'm going to tear apart the ac wiring and just try to hook up one at a time. Once I get it behaving I'll reinstall in my j-box.

Endless fun.

Scott

[Stuff-Builder]

Well, I had a look at my outlet tester this morning, and sure enough it's rated at .3 Watts. So the light was being powered up by the leakage through the solid state relay. That one would have stumped me forever.

I did some testing on one individual switch and everything seems to be behaving properly now. The relay stays off when it should, turns on when it should and the contacts don't close when I shut the computer down. So so far so good. I'm going to rewire the ac outlets tonight, and hopefully put this thing to bed.

Scott