Back EMF - flywheel diode

[leonel]

Hi,
In attachment you can see a schematic to drive a dc motor with pwm clockwise and counter clockwise. The labels OUT are coming from a PLC output transistor and are driving an electromecanic relay (K5) and solid state relays (K6 and K7).
Direction 1 -> OUT2=0, OUT4=0 and OUT3=1
Direction 2 -> OUT3=0, OUT2=1, OUT4=1 (after 500ms from OUT2 turns on, since electromecanic relay are slower than solid state relays).

My PLC program takes care regarding the outputs and prevents that short circuit happens…

My doubts:
Should I put flywheel diodes?
If yes, where?

What happened (twice) is that K7 relay “borned” when I sttoped coming from Direction 1 - > I’m afraid this is related with back emf.

Thanks for your support.

 

[alec_t]

If K7 pulls in before K5 (which is slower) then there is a short circuit, which is a possible cause of K7's failure; but your 500ms delay ought to prevent that. Likewise if K6 pulls in before K5 has dropped out there will be a short circuit. Do you have a delay to overcome that?
Did you ensure the motor had fully stopped before changing direction?
Flywheel diodes (reverse biased) go from each motor terminal to the 24V rail and to the ground rail. The SSRs may already include them.
Why is '5.0' shown on the ground rail?

 

[leonel]

Hi,
Yes, i prevent K6 and K5 with also a 500ms delay...

So the circuit is something like in attachement!? I also thought that SSR could include them...
Please be aware that i'm not sure that the problem is related with back EMF... do you agree it could be the problem?

The '5.0' is related with schematic labels (it's coming from sheet 5).

Thanks.

 

[alec_t]

something like in attachement?

Yes.

I also thought that SSR could include them

You would have to check the datasheet for your specific SSRs.

do you agree it could be the problem?

It could be. But there are, of course, other possibilities such as sticking relay contacts, surge currents, motor acting as generator if not halted before reversing, ..........

 

[MikeMl]

How long does it take the motor to coast to a stop (regardless if you are changing direction or not)?

 

[KeepItSimpleStupid]

If that's a DC motor, your making it way to complicated. Why not use (direction DPDT) and (not enable - SPST) relays?

DIRECTION never has to switch under load and shouldn't.

 

[jamie_s]

you could achieve the same with 2x spdt relays, where both NC relay terminals are connected to GND, both NO terminals to your + rail, and the motor connection one to each relay COM.
Activating each relay will cause the motor to go in one direction. when relays are deactivated the the motor is grounded on both wires which also acts as a brake.
However a motor being an inductive device, i would not really use relays, but use silicon instead

 

[leonel]

Hi,
So, with your comments i can assume the "new circuit" in attachement (with the components i already have) is better than the older one... with that circuit, i do not have a brake fucntion as jamie_s adviced, because 2xspdt electromecanic relays cannot act as PWM - i do not have 2xspdt SSR relays.

Although if the problem is Back EMF i will continue to have problems, no matter what implementation i do... in attachement you can see also the SSR relay schematic - i assumed it should prevent back EMF!??? Or probably not... Even though, should i put the diodes to Vcc and to GND in that circuit?

Thanks for your support.

Best regards

 

[KeepItSimpleStupid]

Brake can be done with the (direction) and (NOT ENABLE). When the SPDT relay of (NOT ENABLE) is at rest the armature is shorted. I used this system on a toy crane I built.

Diodes should always be across the relay coils to protect the drivers.

You can also use an "H-Bridge" IC. Some will give you brake/coast, fwd and Rev.

 

[alec_t]

I'd add the diodes to be on the safe side.

 

[leonel]

Sorry, but i'm a "little" confused... should i put the diodes across the relays coils or across motor Vcc and motor GND (like the h-bridge implementation) ??

The use of H-bridge IC it's not possible, due to not use of pcb, and ssr relay that do brake/coast, fwd and rev and expensive...

 

[alec_t]

should i put the diodes across the relays coils or across motor Vcc and motor GND (like the h-bridge implementation) ?

Both (to protect whatever drives the relay coils and to protect the SSRs).