I just wish it was warm enough to rain! The 555 timer is part of a control circuit and the 555 stops the motor when a coupon has been dispensed. There is a light sensor that detects the end of a coupon on the roll. The motor is restarted by the operator when a coupon has been picked off.

Still need more on the whole mechanism, so far, most has been guess work!
First you say a 555 timer controls the motor then that the operator re-start it?
Also is the motor driving the drum of coupons? If so then the surface rate of the coupons will vary as the drum gets smaller for a given 555 time.
Max.

 

I am working on a machine to dispense paper coupons. It uses a 12 volt DC motor to dispense the coupons off a roll. I need the motor to stop quickly each time a coupon is dispensed so I am using a dynamic braking to stop the motor. The motor cycles on and off according to a signal from a 555 timer. A typical cycle would be run motor for 2 seconds - stop for 5 seconds. This machine could run continuously for hours. I have found two different ways to set up a circuit to brake the motor. One uses a relay and the other a pair of mosfets. They are shown on my sketch. I would like to eventually make these machines and sell them. I am wondering if there is a good reason to use one circuit over the other. There isn't much difference in cost between the two. Is one circuit more reliable than the other or are there perhaps other reasons to pick one over the other.
View attachment 168618

You need 2 MOSFETs: the usual pull down to drive a motor hanging from Vcc, and a second one that just shorts out the motor when you want it to stop quickly. You need to make sure that both MOSFETs never conduct simultaneously..

Many SMPSU controller chips have the; "dead time control" that would achieve this - but they usually drive the MOSFETs via an isolation transformer get at the dead time logic., and you may not be able to get at the dead time circuitry.

Many manufacturers publish an internal schematic - so you can study the logic and re jig it to do what you want. A P-channel brake MOSFET makes things easier unless you have a spare rail higher voltage than Vdd/Vss.

 

First you say a 555 timer controls the motor then that the operator re-start it?
Also is the motor driving the drum of coupons? If so then the surface rate of the coupons will vary as the drum gets smaller for a given 555 time.

The motor turns a pinch roller that advances the coupons. The 555 is only there to stop the motor when a coupon is at the right spot. The machine operator removes the coupon and restarts the motor. There is a dedicated person operating the machine. It isn't operated by random people such as in a supermarket.

 

The motor turns a pinch roller that advances the coupons. The 555 is only there to stop the motor when a coupon is at the right spot. The machine operator removes the coupon and restarts the motor. There is a dedicated person operating the machine. It isn't operated by random people such as in a supermarket.

You still need dead time gating between the 555 and current switching because the 555 won't do that for you - in fact the bipolar 555 has totem pole output shoot through of its own.

 

You still need dead time gating between the 555 and current switching because the 555 won't do that for you - in fact the bipolar 555 has totem pole output shoot through of its own.

So you are saying that with the n-type and p-type mosfets, there can be some small period of time when both are on? I hadn't considered that. Would it be unwise to use the circuit I sketched without the dead time gating? What might happen without the gating?

 

So you are saying that with the n-type and p-type mosfets, there can be some small period of time when both are on? I hadn't considered that. Would it be unwise to use the circuit I sketched without the dead time gating? What might happen without the gating?

Shoot through is a potential risk whatever mix of channel polarities.

you need a small amount of "glue logic" to make sure the on MOSFET switches off before the off MOSFET switches on - otherwise you have 2 MOSFETs in series across the rails pretty much shorting them out for a brief glitch - that can mess up other stuff running from those rails by causing spikes.

 

So this adds a new complication for the mosfet option. I am not sure quite how to add the dead time gating feature. Maybe the relay option would be simpler for me. I will have to study it more.

 

I am freezing today in Canada (1.4 degrees F). The wind is from the south (from the US).
As usual, The Weather Network says it is snowing but actually it is sunny.

 

I am freezing today in Canada (1.4 degrees F). The wind is from the south (from the US).
As usual, The Weather Network says it is snowing but actually it is sunny.

I think the cold has affected you and you strayed off course/off thread a bit?
By the way. Canada is C not F and 1.4°F is positively balmy to some of the rest of Canada.
Max.

 

So this adds a new complication for the mosfet option. I am not sure quite how to add the dead time gating feature. Maybe the relay option would be simpler for me. I will have to study it more.

No less true for bipolar - in fact the CMOS 555 claims no shoot through on its totem pole output. You'd still need to prevent it on the power devices though.

Not preventing shoot through may not cause any damage in that application, but it could contaminate the supply rails with spikes that cause trouble elsewhere. Hoping to get away with it is dodgy design practice.

 

I have been away for a while but maybe I can restart this thread to figure out how to avoid the shoot-through problem with the MOSFET pair. Looking around the internet, I came across this http://www.eevblog.com/forum/projects/would-this-h-bridge-work-without-shoot-through/. Using this idea, I wonder if I might just add a gate resistor in front of one or both MOSFETs to cause a delay. It seems like a simple solution but I wouldn't know how to tell if it works.

 

I am not sure that link works. I will try again. http://www.eevblog.com/forum/projects/would-this-h-bridge-work-without-shoot-through/

 

The gate resistor forms a lowpass filter with the high gate capacitance of the Mosfet for its delayed turn on, and the diode turns it off quickly.
An oscilloscope will show a momentary short circuit on the power supply during shoot-through.

 

I wonder if this circuit I drew up would allow the motor to brake without shoot-through problems?
circuit3.jpg

 

Only if you can find both the PNP and NPN mosfets.

 

I was going to use an IRF9540 P-channel and an IRF520 N-channel. Is there something more to the selection?

 

Gerstley, I'm afraid you're going to need something a bit more elaborate than what you've drawn. Mosfets do not behave as ordinary bipolar transistors, and they need to be placed and driven in specific ways. Also, even if your circuit could work (it won't) there would still be overlapping signals between fets as they're being switched on and off, with undesirable effects.

Right now I gotta get busy, but I'll be back tonight with specific advise and observations.

 

I was going to use an IRF9540 P-channel and an IRF520 N-channel. Is there something more to the selection?

They just like any mosfet aren't PNP or NPN, those are types of transistors, not mosfets.

 

Must confess, I don't know why I would want to dispense two seconds worth of paper every five seconds.

@cmartinez Hadn't considered that as an option. Excellent point.

Bogroll dispenser for someone with a dose of the trots?

 

Right now I gotta get busy, but I'll be back tonight with specific advise and observations.

I would very much appreciate your thoughts whenever you find some time. I am wondering if the relay would be a simpler option?