[MadeInTheShed]

Ive spent the last few evenings trying to set up a relay to control my router and vacumm on/off's.
Here is the set up........

Signal from Hobbycnc board (via lpt1) turns a 5v solid state relay off and on. Works fine.
As this relay cannot deal with the current that the motor and vaccumm draw, I have this 5v relay switching a 12 v supply , works fine.
Connect this 12v to my 30A relay (which will switch the AC to the router) and it will turn on via Mach 3 control, but will not turn off. If I disconnect the relay, it switches off, reconnect it and it stays off, till mach 3 tells it to turn on again.
If I use a meter to measure the voltage, instead of connecting the relay, it works fine - Mach 3 ON, 12v ....... Mach3 OFF 0v.

Any ideas why this is happening?

5v and 12v supplies are being sourced from a spare connector from the Mach3 computer.

Ive included the schematic to try and make it clearer..

[mkc2]

Hi
I did that same thing and had problems,found out i was using the wrong relay.
I also used a 5volt relay on my interface board to switch a 12v relay to turn on 240v router. wanted to keep the 12v relay and 240v away from the interface card so its interferance free.

This is what i ended up doing. made a rough sketch of your drawing sorry !! hope you can understand it.

Michael

[Al_The_Man]

You are over complicating things, all you need is a 2N7000 and a relay.
There is no point using a SSR if it switches a mechanical relay, you have the isolation via the relay.
Do a search here for 2n7000, there are a few examples.
http://www.cnczone.com/forums/showth...ghlight=2n7000
If you want to use a SSR with AC output to switch a spindle direct use a 2n7000 directly off the port and the SSR would be connected to the drain of the Transistor.
The beauty of the 2N7000 is that it was made for this and it does not load the port up like a bi/polar transistor or Opto.
Al.

[MadeInTheShed]

Thanks for your replys. gives me more to work with.
I think I have a ULN2803a somewhere, I will give that a try.

[Al_The_Man]

They also work fine, they have the reverse EMF diodes built in.
I should have added that a SSR may required if using a contactor in place of a relay, due to the higher coil current.
Al.

[MadeInTheShed]

Attached is an image of the circuit I am using for the ULN2803A.
Obviously, The pin numbers on the port are different (using pin 1 on the parralell port to pin 1 on the ULN2803a), but the rest of the circuit is identical.
Such a simple circuit, yet its not working!
If I measure V at Pin 1 on Parallel Port, nothing connected, signal sent, I get 4.8V. Connect the ULN2803a as per schematic, and same settings I get 3V.
V at pin 18, is 0.56V whether 0V or 4.8V is present at Pin1.

Any ideas?

Crook Chip maybe?

[Al_The_Man]

Pin 9 should be common to the parallel port common and VCC common.
If VCC does not originate from the PC but is a separate supply, this is most likely the problem.
ULN2803 is not an isolator.
Al.

[MadeInTheShed]

VCC is 12v, from a spare connector out of PC PSU.
I have also tied (with a jumper lead) Pin 9 ULN2803, Pin 25, and VCC common.
Same results.

I have a ULN2003 as well, might chuck that one in and see if I get similar results

[Al_The_Man]

If the 12v is originated from the same PC then it is still preferable to connect a common from the spare PC connector to the IC common as well as the 12v+, this avoids the relay current flowing through the port common back through the board.
Al.

[MadeInTheShed]

Still not sure whats going on here, tried a new ULN2803a, and got similar results except now I get 0v/3v from pin 18 of ULN2803a, when pin 1 is toggled, 12v to pin 10.

Going to chuck this in the bin and try the 2n7000 now. What diode should I use across the coil?

If that doesnt work, I'm going to give up!!

[Al_The_Man]

The output pin 18 should be 12v when pin 1 is zero, and if you apply a signal above 3.5v to pin 1, then pin 18 will be close to zero WRT common.
Try a jumper from the 5v PC supply direct to pin 1 to test the zero to 5v input.
The output should switch.
A 1N4005 or similar is sufficient for coil diode.
Al.

[Torchhead]

The first circuit did not work because an SSR is an AC switch. In side it's basically a Triac (two SCR's) which latch on when the gate goes positive and only turn off if the gate drive is removed AND current to the load passes through zero (every 8ms on a 60HZ wave). DC just locks them on until you remove the DC load.

TOM Caudle
www.CandCNC.com

[MadeInTheShed]

The first circuit did not work because an SSR is an AC switch. In side it's basically a Triac (two SCR's) which latch on when the gate goes positive and only turn off if the gate drive is removed AND current to the load passes through zero (every 8ms on a 60HZ wave). DC just locks them on until you remove the DC load.

TOM Caudle
www.CandCNC.com

Ah-ha!
That explains it. I'll have to find another use for this relay now.

[MadeInTheShed]

The output pin 18 should be 12v when pin 1 is zero, and if you apply a signal above 3.5v to pin 1, then pin 18 will be close to zero WRT common.
Try a jumper from the 5v PC supply direct to pin 1 to test the zero to 5v input.
The output should switch.
A 1N4005 or similar is sufficient for coil diode.
Al.

Thanks Al. Not sure if I have tried that yet, done that many tests!! I'll try that tonight again.
Just picked up some 2n7000's while I was out to try that method as well.

[MadeInTheShed]

Interesting.........

I tried with the 5v test (from the PC powersupply) on the ULN2003A and got same results with the meter.
I dont know why, I then connected the relay, and lo and behold it works !! Relay clicks on and off with Mach3 control.
I measured again, and still get the same results as I have been getting, so I guess I must have been using incorrect reference points for my meter ground.

Thanks for all the advice.

:cheers:

[Al_The_Man]

If you were measuring the output without the relay connected it will always be around zero as the output is open collector.
It requires a load connected to see any kind of voltage when off.
Al.