DC-DC relay help... no mohawks allowed:)

rjo__

I've been looking at this on and off for several weeks and have come up empty.

I am driving two (at the moment... 4 in the future) 24V 15A wheelchair motors with a Sabertooth 2X60 control board from DimensionEngineering..http://www.dimensionengineering.com/products/sabertooth2x60

I prefer using analog(pwm) mode... which requires a 2.5v signal to be neutral. Any other signal (or no signal) causes the controller to drive the motors.

I have a circuit, which at start up maintains a 2.5v signal until the circuit is turned off by a Propeller pin output. Theoretically, if the Propeller fails, the circuit will stop the motors.
And until the Propeller starts up... the circuit remains active, keeping the motors neutral. So far, so good.

In actual practice, I also want to use some type of normally open relay...also hooked up to a Propeller output... which would only allow the motors to receive current once engaged by the Propeller.

I have some DC-DC SSRs... but there is a problem. This particular controller features regenerative braking... and from what I read, if I simply put an SSR between the battery and the controller...
and the relay is turned off while the motors are turning, the controller could be damaged. I have read that this could occur, while simply pushing the wheelchair along.

However, since the motors are bi-directional... I can't put the SSRs between the controller and each motor... can I?

Assuming I can't use SSRs what is the best mechanical relay to use between the controller and the motors... and can be controlled by a Propeller?

Thanks

LoopyByteloose

I'd forget the solid-state relays as they create quite a bit of heat, may need heat sinks, are SPST logic, and would be much larger than good 20 amp 24VDC mechanical relays.

Having said that, you really need to have a logical plan for all you want to do with a clear set of fail-safe priorities.

The Propeller's 3.3 VDC can be boosted to 24VDC to engage the relay coils without much difficulty. You could use either a high gain transistor, a power darlington-pair, or a power MOSfet to drive the relay coils. I think a TIP12O power darlington would drive just about any large 24VDC relay coil you might use, and there is very good likelyhood that somethig much less, such as a simple 2n2222 which can provide 500ma at 24VDC would be adequate.

But this all takes knowing what the power requirements of the selected relay are and working backwards. So you do have to figure out what switches you want where and what sequence of control you want. Also you need to understand how they respond to having the power completely removed.

So having a clear set of priorities and working logically with how the devices power up and down is the first issue. Then there are whatever logical controls you want to impose.

Why not put the lock out relays .. which would be DPDT .. between the motors and the controller? When all power is removed, the controller would see open circuits instead of motors. And motors that freewheeled would be sending power nowhere. Of course, it you need brakes, you will have to find them via another mechanica configuration. But I hope you weren't considering the regenerative braking as your sole source of brakes in a real world situartion.

Any robot that is large and heavy and going to roll up and down slopes needs real brakes.

rjo__

Exactly. I am trying to find a good mechanical relay to place between the motors and the controller for safety. Based on my understanding, I had eliminated SSRs. But I also wanted to check my understanding.

I think my problem is that the specs are confusing me:) I need to pass up to 25A peak at 24V. But most specs are for AC... and I just don't get it:(

The brake issue. You are right. There was solenoid style brakes on the motors, but to fit the encoders on the motor shaft, I had to remove it. I have some other wheelchairs, where this is not a problem.
This is a test platform, so I think I can get away with controller braking for the moment.

Any help is appreciated.

Rich

kwinn

In theory an AC SSR should work in this application. The only catch is that the SSR would not turn off until the motor came to a stop (no current flowing in either direction). Mechanical relays typically have a much lower DC than AC voltage rating because the AC voltage crosses 0 volts 60 times a second which reduces arcing quite a bit. For your application a 50A 240 volt relay should work. Another avenue to consider would be a truck (tractor trailer) starter relay.

rjo__

Thanks kwinn... exactly what I was looking for. If anyone has a favorite vendor, speak now or forever hold your peace:)

LoopyByteloose

The are essays and a body of engineering knowledge about reducing arcing on relay terminals. For anything over 5 amps, one can add circuits to snub some of the damage.

You may find that a 50 amp DPDT 24VDC relay is difficult or expensive to locate and may have to settle for a more economical 30 amp.

stamptrol

A good source of high current, low voltage relays is the trucking industry. Usually sold to switch auxiliary lighting. A standard relay I use for motor isolation is rated 40A on the NO contact/30 A on the NC contact, coil is available in 12v and 24 volts. The package is very small ( 1" x 1" x 1.5") and uses push on connnectors. About $10 each.

In your application, it will allow you to establish the control voltage from the Prop before energizing the power circuits. This will prevent unwanted operation of the motors and will give an independent means ( Emergency Stop switch) to shut the motors off as well.

Cheers,

rjo__

This is going to be a good day.

LoopyByteloose

Try something like this, but notice that you are going to actually test the coil to get how many milliamps are required to activate the coil.

Coil current rating at 2VA is a bit confusing. It is a 24VDC coil. I would want coil resistance in ohms.


And while the contacts are rated for 240VAC, 10% of that is not uncommon for use with DC.

http://www.grainger.com/Grainger/DAYTON-Power-Relay-3X749

or

http://www.galco.com/buy/Magnecraft-Schneider-Electric/W199X-13

LoopyByteloose

Take a look at Wikipedia's Relay article and the Derating section.

For use with a motor, the article suggest 20% of the rated use... so a 40 amp relay would be used for an 8 amp motor load.

So ideally a 25 amp motor would want a 125 amp rated relay. There are 130 amp 24VDC automotive relays, but they seem to be single pole single throw, single pole double throw, and double pole single throw configurations. I can't seem to locate a DPDT.

http://www.onlinecomponents.com/te-connectivity-amp-brand-1-1414428-0.html?p=28218728

Search 'automotive relays high current devices'

BTW, the 240VAC 40 amp relays are acceptible for up to 28VDC at 40 amp rating. This is another derating factor by use, not by voltage.

http://en.wikipedia.org/wiki/Relay#Derating_factors

and then there is the issue of contact arc suppression

http://en.wikipedia.org/wiki/Contact_protection

http://www.industrologic.com/mechrela.htm

rjo__

A very, very good day.

Thanks guys

kwinn

Loopy Byteloose wrote: »

........

Coil current rating at 2VA is a bit confusing. It is a 24VDC coil. I would want coil resistance in ohms.

................

This is why we learned E=IxR and all it's derivatives.

2VA / 24V = 0.083A

24V / 0.083A = 288 Ohms

LoopyByteloose

And so, the transistor driving the coil at 24 volts would have to provide at least 83 milliamps.

Since a 2n2222 is good to about 800 milliamps and 40Volts, it has enough headroom, but would require more current that a small darlington solution.

I guess the coil says VoltAmps because one might power the coil at 20 volts, and then 2/22 = 0.100A or 100 milliamps... while resistance would become 220 ohms. Not really clear on the implications of resistance being different.


Volts x Amps = Watts, so is it really the coil just wants to be driven at the right amount of watts and not limited by physical resistance?

Driving the 2n2222 at 3.3 volts and 10ma would likely work well... so 3.3/.010 = a 330 ohm current limiting resistor between the Propeller output and the transistor base.

kwinn

Loopy Byteloose wrote: »

And so, the transistor driving the coil at 24 volts would have to provide at least 83 milliamps.

Since a 2n2222 is good to about 800 milliamps and 40Volts, it has enough headroom, but would require more current that a small darlington solution.

A 2N2222A has a minimum Hfe (gain) of 35 so the base current needed to provide 83mA would be 0.083 / 35 = 2.371mA, which would require a base resistor of a little over 1K. I usually use a gain of 20 for the resistor calculation to be sure the transistor is saturated, and that would give a base resistor of 626 ohms, so I would use a 680 ohm base resistor in this case.

I guess the coil says VoltAmps because one might power the coil at 20 volts, and then 2/22 = 0.100A or 100 milliamps... while resistance would become 220 ohms. Not really clear on the implications of resistance being different.

Not really sure what you are trying to say here. The coil is designed (# of turns & wire size) to have the resistance needed to limit the current to a value that will reliably pull in the relay contacts. The VA rating is based on the current drawn at the rated voltage. If you powered the coil with 20V the coil would draw less current (20 / 288 = 69.4mA) and might not pull in the relay contact. Of course the actual VA drawn would only be 20V x 0.069A = 1.38VA

Volts x Amps = Watts, so is it really the coil just wants to be driven at the right amount of watts and not limited by physical resistance?

For a coil powered by a DC voltage it is watts. The VA rating is how the manufacturer chose to rate it. It could also be rated as 2W at 24VDC or 188 ohms, 24V DC.

Driving the 2n2222 at 3.3 volts and 10ma would likely work well... so 3.3/.010 = a 330 ohm current limiting resistor between the Propeller output and the transistor base.

See the first answer. A 330 ohm resistor would certainly work well, but it would provide more base current than is really needed.

LoopyByteloose

A. Yes, you could drive the 2N2222 with less current. But there are pro and con arguments of driving the device to saturation versus driving it to linear operation. And we do know that gain figures vary widely in actual use. I have no problem with using a 1K resistor and demanding less of the Propeller i/o if it will work. I actually prefer to have the i/o on the Propeller provide as little power as possible... to make the Propeller last longer.

B. I hate the VA specification as it seems to be going in the back door when all your really need to know is how many Amps are needed at the proper voltage. I do admit that the coil has reactance as well as resistance. So it actually may require more current that what the measured resistance would indicate. I just don't see why they give me a specification that requires obscure math calculations.

C. I am deeply concerned with the current derating of mechanical relays to 20% when used with motors. This is a huge number and in the case of a 25 amp motor, it results in a jump form a $40 DPDTrelay to maybe having to use more than one relay DPST costing over $100USD.

kwinn

@Loopy

A. For switching a relay on driving the transistor into saturation is a good thing since it minimizes the transistor power dissipation. For times when maximum switching speed is wanted it is not so good.

B. I agree. The most useful rating would be the coil current and the nominal AC and DC voltage.

C. This is done mainly to deal with the motor starting current and potential arcing on stopping.

LoopyByteloose

Regarding C and the derateing..

These seems to take me back to looking at Solid-State Relays as there is NO arcing and the newer ones are smaller, cooler, and much cheaper that the mechanical relay solution.

Solid State may be much cheaper to build as well.

But to get a DPDT relay in solid state, one would have to use 4 SPST devices.

And, the hazard of solid state relays is that they fail to an ON state, while the mechanical fail to an OFF state.