I've been messing around with some SSRs lately for a project I'm working on at my laundromat. Right now I have mechanical switches on my coin acceptors. They work well enough but I wanted to play around with a different type of coin acceptor that doesn't use them. This one uses a -8VDC pulse when it gets a coin. It also has the option of being able to accept a second type of coin and you can set the number of pulses. Example one pulse for a $.25 coin and 4 pulses for a $1.00 coin. The electronics in the machine aren't compatible with a -8DVC pulse so I have the pulse going to the - coil on a mechanical relay and the wires that went to the old mechanical switch hooked up to C and NO. It works well enough when you do the $.25 with one pulse, but the 4 pulses for $1.00 come in too fast for the mechanical relay to process and it only credits for 3 pulses or $.75. Tried using some off the shelf SSRs but they only work for switching AC loads. Can't find one for analog signaling. Did some research and am going to try to build with TI TS12A4515PE4 analog switch IC. Should work fine in this application because of the low current involved, but I was wondering if any of you guys knew how to build a SSR that can handle heavier loads of say 5A instead of the 30mA these ICs can handle.

Problem with off the shelf SSR is that they appear to need a load to function correctly. What I'm looking for is a direct relay replacement. Something that gives continuity as soon as the "coil" is energized and takes anything on the output terminals out of the equation.

 

Are you sure that it is the relay that is too slow and not the machine itself that can't cope with the speed of the pulses.
You don't say what voltage and current the contacts on the relay are required to carry and if it is AC or DC. You mention 5 amps but that seems rather high for a control signal. Also what is the time of the spacing of the 4 pulses ?

Les.

 

LCA701
LCA710

These are small solid state relays based on MOSFETs rather than TRIACs, and should be fast enough.

Digi-Key shows over 400 part numbers for things that could work.

https://www.digikey.com/en/products...gZRMGsTlGQth+gAtGY0AREGMRlPEsCdEsYAcvANiBDRQA

ak

 

As AK noted, an SSR with a MOSFET output (DC output) will work without any particular load on the output.

 

Does the diagram below represent your current setup? A & B go to the machine electronics...



Can you say what voltage is across A & B when the contacts are open, and the current through them when closed? Also, which of A or B is connected to ground?

What relay are you using?

The LCA710 that AK suggested should work fine...

 

Are you sure that it is the relay that is too slow and not the machine itself that can't cope with the speed of the pulses.
You don't say what voltage and current the contacts on the relay are required to carry and if it is AC or DC. You mention 5 amps but that seems rather high for a control signal. Also what is the time of the spacing of the 4 pulses ?

Les.

Could be the machine, but I'd like to try the relay first. The Board on the machine runs on 12VDC and the lead out to the switch on the coin mech looks like +8VDC but it could have been because I have a bad ground. I imagine that it wouldn't need much current to work. The 5A was for a more general purpose relay for other uses besides this.

 

Does the diagram below represent your current setup? A & B go to the machine electronics...

View attachment 281672

Can you say what voltage is across A & B when the contacts are open, and the current through them when closed? Also, which of A or B is connected to ground?

What relay are you using?

The LCA710 that AK suggested should work fine...

View attachment 281674

Sadly I'm running late for work or I'd make a drawing. The old coin mech had a mechanical NO switch that got hit every time a coin dropped past it. That would complete a circut to the coin board and give credit for the coin. The board has a 220V input but it's stepped down to 12VDC. Output I was getting from one of the leads on the old switch was 8VDC but I could have had a bad ground. The two leads of that switch were moved to an Altronix RB5 relay terminals C and NO. The new coin mech has its own power supply (old mech was mechanical didn't need power) and every time a coin passes through it it outputs a -8VDC pulse. The coin mech and the relay both use the same 12VDC power supply. I didn't connect the grounds of the board to the ground on the relay and coin mech because I wanted to keep everything isolated. When the new coin mech detects a good coin it outputs -8VDC to the neg side of the coil on the RB5 and trips the relay. I'll see if I can draw a picture after work. I appreciate the help.

 

My guess would be that the relay that is used on the "Altronix RB5 relay module would be fast enough. It's pull in time is 15 mS max and its opening time is 5 mS max. (These figures are without a back EMF clamp diode and there does not seem to be a diode fitted to the Altronics module.)
Here is a link to the data sheet on the relay.
We will need more data from you to be SURE that the relay is fast enough.
A link to the datasheet on the coin acceptor would be helpful as that will probably give information on the speed of it's output pulses.

Les.

 

Not sure which coin mech the TS has but looking around the web there seems to be a fairly common interface with 1 - 10 pulses per coin, 30/50/100mS wide pulses. As far as I can see the mark/space ratio is fixed, so 30mS pulse is 30mS HI, 30mS LO. HI & LO are configurable as NC or NO and the output is open collector needing a pull-up resistor from 'signal' to V+ (12v) or an external supply eg 3.3v or 5v.

Not sure where -8v comes from; suspect its somehow measured relative to +12v rail?

 

Not sure which coin mech the TS has but looking around the web there seems to be a fairly common interface with 1 - 10 pulses per coin, 30/50/100mS wide pulses. As far as I can see the mark/space ratio is fixed, so 30mS pulse is 30mS HI, 30mS LO. HI & LO are configurable as NC or NO and the output is open collector needing a pull-up resistor from 'signal' to V+ (12v) or an external supply eg 3.3v or 5v.

Not sure where -8v comes from; suspect its somehow measured relative to +12v rail?

Sadly the datasheet on the coin mech doesn't have much data. It's designed to run of 24VAC. My experience with anything electronic that "runs on AC" is that it actually runs on DC internally and it converts the AC to DC. The coin mech (Slugbuster II) runs just fine with a 12VDC input to the yellow and black wires. It outputs -8VDC pulses to the blue wire. That's the one I have hooked up to the - coil on the RB5 relay. One thing I failed to check that could be a factor is if the original coin mech switch was NO or NC. That could be a factor in it getting only 3 pulses when it's supposed to get 4. Could be a noob mistake I'll check it out with my meter the next time I'm at the laundromat.

 

Sadly the datasheet on the coin mech doesn't have much data. It's designed to run of 24VAC. My experience with anything electronic that "runs on AC" is that it actually runs on DC internally and it converts the AC to DC. The coin mech (Slugbuster II) runs just fine with a 12VDC input to the yellow and black wires. It outputs -8VDC pulses to the blue wire. That's the one I have hooked up to the - coil on the RB5 relay. One thing I failed to check that could be a factor is if the original coin mech switch was NO or NC. That could be a factor in it getting only 3 pulses when it's supposed to get 4. Could be a noob mistake I'll check it out with my meter the next time I'm at the laundromat. View attachment 281766

A note on the -8V pulse. I haven't actually be able to measure it with my meter because it's too fast. I'm just going by what the data sheet says. All I know is that it is enough to trip the RB5 and it will only trip when hooked to the - side of the coil.

 

Here is a link to the datasheet on the slugbuster 2
It says that it gives an 8 volt pulse out on the blue wire BUT it does not say what that is with respect to.
As it marks the black wire as common/ground I would assume the 8 volt pulse is with respect to that wire.
As the black wire is common the the AC input the device must only use half wave rectification. (So that one side of the AC input is the same as the common of the DC output.
kraljsved
Can you confirm that this is the case ? Can you also post pictures of the printed circuit board (Component side AND etch side.) in the slugbuster ? You seem to be treating the red wire as common which could explain why you see a negative pulse.

Les.