Help needed choosing relays

[moorea21]

Hi,

I have a treadmill motor (rated 1.5Hp, 4800 rpm, 230V DC) and controller which I can run from a picaxe pwm. The controller only runs in one direction, but I want to use it in reverse too. In my intended application, it will always stop for at least a fraction of a second before reversing, so I could achieve this using 2 spdt relays (as attached, also here https://quasarelectronics.co.uk/media/ecom/images/reversible-dc-motor-using-2-relays.gif), on the power leads to the motor, controlled from picaxe output pins. The whole device is for microprocessor controlled lifting via a screw jack basically, but it's operation isn't critical safety wise, ie nothing physically dramatic would happen if it failed.

The relays shouldn't be a problem electrically if grafted into the existing wiring, although I supect I may only find relays with 12V coils suitable for this kind of load?

Which brings me to my question. How should I calculate the Voltage/Current rating needed for the relays? Any suggestions as to where to buy something suitable? I'm thinking Farnell, mouser, etc etc, or ebay if the quality isn't an issue there.

Thanks

 

[hippy]

I have a treadmill motor (rated 1.5Hp, 4800 rpm, 230V DC)

Not sure that actually is 230V DC - Perhaps a typo for 230V AC ?

could achieve this using 2 spdt relays ... on the power leads to the motor,

I am not convinced that's a good design. When the motor is switched out of circuit, that is achieved by shorting both sides of the motor together.

I know such things can be done to brake a motor but not sure it's that simple. It doesn't feel right to me but it's not my field of expertise.

I would have done it using two relays; a DPDT to switch the motor between L and N, and another to connect and disconnect at least L -

         /         / o-------------.
L ------O O------O/:               |
         :         : o---.         |
         :         :     |         |
         :         :     }---[O]---{
         :         :     |         |
         /         / o---'         |
N ------O O------O/:               |
         :         : o-------------'
         :         :
      On/Off    Direction

 

[Aries]

Can you not use one DPDT instead with (as Hippy says) another to switch on and off? I have done so with a much smaller motor with no problems. In the past, I have used relays from RelPol. The project actually used the inrush version (which is SPDT) and it has worked well for over a year, on a daily basis. This is the sort of relay you might use:

https://www.tme.eu/en/details/rm84-2012-35-1005/miniature-electromagnetic-relays/relpol/

 

[moorea21]

Thanks both,

I originally wanted to use spdt's as I already have some, but they have 12V coils and are 240V 10A, which I started to think were not really substantial enough(?), I could use a 5v coil dpdt instead. I'll look for something similar to the RelPol one in the UK; suggestions for a suitable source are welcome, in case I cheap out and buy something from ebay that burns my house down... Re-drew Hippy's diagram too, to be certain I've understood it.

 

[hippy]

Re-drew Hippy's diagram too, to be certain I've understood it.

Yes; that's how I intended. The On/Off can be DPST, or DPDT, even SPST on L through I think it's always best to disconnect L and N if possible.

 

[moorea21]

I would think it would still work without the relay on the left? The voltage across the motor when it isn't turning should be 0V, so its the same as having the on/off relay?

Would 230DC 8A be sufficient for the dpdt?

Thanks Hippy

 

[hippy]

I would think it would still work without the relay on the left? The voltage across the motor when it isn't turning should be 0V, so its the same as having the on/off relay?

You would need something to disconnect the power, relay or switch.

If you used two separate relays, one for L one for N, wired as I'm suggesting, to remove voltage from the motor the relays would short L to N and that would be a disaster.

Would 230DC 8A be sufficient for the dpdt?

No idea. It would depend on the motor specification, in-rush current.

One Horse Power is apparently 746 W, so 1.5 HP would be 1119 W. For a 230V DC motor that would be about 5A but I'm sure there's some 'fudge factor' to include for AC and I'm guessing that's normal running, not in-rush.

 

[inglewoodpete]

Due to the reactive load, you will need to use an over rated relay or contactor to prevent contact welding resulting in locking of the relay. How do I know? Bitter experience!

 

[moorea21]

So in view of the fact that I'm unlikely to be able to get the full specs for the motor, including things like start up current (is that the same as in-rush current?) it may be necessary to use a factor of safety of, say, 2, when specifying the relays, ie 10A? And maybe have a 'ramp up' of motor speed after starting quite slowly. I'd like to learn to weld, but welding relays closed isn't something I would need to practice I think. Thanks for the heads up on that!

 

[Aries]

The RM85 RelPol relay (SPST) has a 20mS inrush current of 80A - you would need four to replicate the effect of one DPDT or two SPDT relays (and you would have to be very careful not to short the wrong ones together). I bought 20 RM85-5021-25-1005 (5 volt coil) in 2018 from TME for about £30 including postage - they were delivered from Poland the next day.

 

[moorea21]

Thanks Aries, that looks like the best option for current and cost. I'm not sure how they would be connected to emulate an spdt and a dpdt as in hipyy's diagram?

Also, I've just noticed in Hippy's original reply :

moorea21 said:
I have a treadmill motor (rated 1.5Hp, 4800 rpm, 230V DC)

Not sure that actually is 230V DC - Perhaps a typo for 230V AC ?

Nope, it's definitely 230V DC

 

[hippy]

So in view of the fact that I'm unlikely to be able to get the full specs for the motor

Google may reveal some useful information there.

From what I've found it does seem it is a DC motor, but odd they state 50Hz/60Hz.

 

[moorea21]

Yes, I wonder if it only ships with its controller/power supply, which is for 50/60Hz. Google didn't help today. Maybe tomorrow...

 

[premelec]

Not sure if it got mention somewhere - DC contact ratings often require quite a bit lower [6x?] current capacity than same contacts on AC - I think primarily because of arcing [with AC current goes to zero between peaks]. If you look at an established relay mfr's website there may be discussion of this.... I just looked some and found forum comments but not any official papers... You might consider switching the AC off while switching the DC end or look for fancy stuff like Kilovac relays ... or make a high voltage H bridge... good luck...

 

[Hemi345]

If your search for a robust DPDT relay doesn't turn up anything, you might consider using two solid state relays. They have a low input voltage capable of handling fairly high power and the PICAXE would be opto isolated from the load:

US Stock SSR 40A DC Control AC 110V Solid State Relay Heat Sink Thermal Grease | eBay

1x heat sink w/ 2 screws. 1x thermal grease. 1x SSR-40 DA Relay. Fine anti-interference feature, fast switching speed.

www.ebay.com

Crydom solid state relays are a solid brand available at Mouser.

 

[tmfkam]

Does the motor have brushes? How many wires are brought out for connecting?

Many 240V motors that can run on DC will run very well from 24V upwards. Sinclair famously used a series universal motor (AC or DC) in the C5 which ran off a car battery (or two). Try powering the motor (with no controller) off load (no gearbox or treadmill) from as high voltage DC supply as you can, (24 - 48V?) and see if it turns.

An AC motor's direction is usually changed by altering the phase of only one (or more) of the windings, relative to another winding, this would normally require more than two external wires for termination.

 

[moorea21]

Hi premelec,

I have noticed that relays rated for high current AC are rated for much lower power DC, which actually suggests the reverve of what you have mentioned, so I'm confused now! Also, what would be the advantage of switching off the AC end of the power supply?

Hemi345:
I have so little experience of relays, I hadn't considered solid state one, and not for any good reason. I'll look into those ones, thanks.

tmfkam:
The motor has a red + wire, a black - wire, and an earth that it shares with the AC side of the power supply. Having taken the motor apart, it's definitely a brushed DC motor, not a universal. The brushes are perpendicular to the shaft, so it should reverse easily enough.

What I may do is buy , say, 20A spdt's with 12V coil, and use them to experiment without using any microcontrollers, just switching a 12V power supply on and off.

General Question:

If my relay turns out not to be up to the current required, what sort of failure mode can I expect? Will it be just a dead relay due to contacts arcing together, or will I end up blowing myself up and being remembered posthumously as the bloke who knocked out the UK's power grid, or something in between?

Thanks for the advice so far.

 

[Aries]

I started to use inrush relays to switch fluorescent lamps, which have a high start-up current. The effect on a non-inrush relay was to weld the contacts so that it was always on. In the case of a lamp, it was a bit irritating, in the case of a motor it could likely be a lot worse. ALSO, if your contacts weld, and you then try to reverse it by switching on the "other" relay, you will at least blow a fuse somewhere.

 

[moorea21]

Thanks Aries. So it would be slightly beyond inconvenient, and maybe lead to a dead power supply. E-stop switches on the limits of the screwjack would disable the motor's power supply when it gets to the end of its travel, so a rougue motor wrecking things should be avoidable.

Another idea I've had would be to use a linear motor from a floppy disc driver to throw a manual reversing switch; it would need to physically go past its 'off' position on its way to the opposite pole, so that would also take care of the 'must stop motor before reversing' problem, if timed correctly. I bit of a bodge though...

 

[PhilHornby]

I have a treadmill motor (rated 1.5Hp, 4800 rpm, 230V DC) and controller which I can run from a picaxe pwm. The controller only runs in one direction, but I want to use it in reverse too.

I found a GMD118-1 motor, listed on Gumtree (of all places). There's a photo showing a motor and a controller board, for those of us unfamiliar with it.

Are you using the controller board that came with the motor? If so, does the motor actually reverse when the supply to the controller is reversed?

When you say "I can run from a Picaxe PWM", have you have got that bit working? Are you feeding it a 230v DC supply and if so, how are you obtaining that? ( schematic would help)

If it is DC, you probably can't use a commercial SSR, because they use back-to-back (side-by-side?) SCRs (that will latch). (Cheapo SSR's, like the ones I buy, use Triacs - but same issue). (If it's a permanent magnet motor, with brushes, the SSR might drop out, as the commutator switches, I suppose)

Also, why PWM? ... are you successfully varying the speed of this motor?

If the answer to the last question is "yes", you might be able to avoid the worst of the transient currents (by reducing the speed to zero, before switching the relay).

Incidentally, with an inductive load, like a motor, you don't get 'in-rush' currents - you get the opposite (outrush?), when the stored energy in the coil gets released). Rather than an over-specced relay, a bit of spark suppression is probably in order.
This might be useful: https://www.illinoiscapacitor.com/pdf/Papers/spark_suppression.pdf (if nothing else, it explains why you need a resistor and capacitor, rather than just a capacitor - which I didn't appreciate). I didn't fully understand their formula for calculating component values though...

 

[moorea21]

Hi PhilHornby,

I'm using the controller that came with the motor, it outputs the 230V DC, and yes, the motor reverses when the contacts are reversed. Speed is altered via a pot (native to the controller,) which I'll replace with a smoothed picaxe pwm signal. I haven't built that yet, but have used similar with other DC motors in the past. I'll probably build this one www.youtube.com watchv=tSK5ZVXY1dA. See attached diagram (nb the wire colours on the pot on my controller are different).

Note, this is for running the treadmill motor as a spindle motor on a milling machine via a PMDX 122 breakout board from a PC's parallel port running Mach 3 cnc software, hence the pin designations on the left of the drawing, but as this signal is just pwm, any pwm signal of 5V should work.

I'll leave SSRs alone maybe, I didn't think in terms of commutators causing unwanted latching problems.

Thanks for the link, that's a good bit of advice, although I'm not sure if just guessing the load current at contact opening will yield a usable circuit design, but it may still work fine if only approximate.

 

[PhilHornby]

So, the plan is to stick with the original controller and reverse its (i.e. the controller's) 230V DC output with something interposed between it and the motor. The controller already performs speed control, via an analogue control - which you intend to interface to the Picaxe via PWM.
(Just making sure I understand )

It struck me, that the normal way of reversing a DC motor is via a "H-Bridge" (something I've never used in my life ).

Most of the ready made modules are intended for much lower voltages than yours, but I came across the PWD13F60 which is rated 8A 600V. (I didn't check the price) See: https://www.st.com/en/power-management/pwd13f60.html
Better still, there is a development board available which would save the aggro of making a PCB for it.

Of course, I have no practical experience of the thing, but from the 'typical application diagram' in the datasheet, it looks you'd connect the output of the controller to the "VS" pins and your motor to "OUT1" and "OUT2" (with some sense resistors on SENSE1/SENSE2 if you want to detect the motor stalling). Then, the direction of travel (if any) would be set by the logic inputs on LIN1/HIN1 and LIN2/HIN2. It also needs a low voltage supply to Vcc1.
The device also protects against illogical control inputs - thus preventing software errors from starting fires

The Picaxe has built-in support for interfacing to H-bridges (which again I haven't used) - but that might make it even easier to use (including the speed control).

I'll leave SSRs alone maybe, I didn't think in terms of commutators causing unwanted latching problems.

Actually, I meant it the other way round ... as in, a brushed DC motor, might just cause a Triac/SCR to drop out, when it switches. It probably depends on the size & duration of the back EMF. I think that is going to be 'appreciable' in the case of this motor!

 

[JimPerry]

https://www.ebay.co.uk/itm/STMicroelectronics-EVALPWD13F60-High-Voltage-Evaluation-Board-Brushed-DC-Evalua/392638210389?epid=28026840197&hash=item5b6b0fa155:g:~t4AAOSwXqpeHaXT £40 Ebay UK

Or £10 for chip

 

[moorea21]

That's great, thanks for taking the time to find that, I had no idea you could get an H bridge on an IC that would handle anything like that amount of power; it wouldn't have occurred to me to look for one. I certainly wouldn't have wanted to build one from scratch if at alI possible. I've ordered one on its development board from Farnell (https://uk.farnell.com/stmicroelectronics/evalpwd13f60/eval-board-half-bridge-driver/dp/2822452) and its half the price of the ebay one. (Ta Jim.)

 

[PhilHornby]

I just noticed something in the data sheet...
"The EVALPWD13F60 is 48 x 53 mm wide, FR-4 PCB resulting in an Rth(J-A) of 18 °C/W, capable to drive loads up to 2 A RMS, without forced airflow cooling. "

You might need a small fan! Perhaps an old CPU, or 'Northbridge' cooler? I wonder if you're supposed to just blow air across the board, or add a heatsink and blow it downwards; it doesn't really say...

 

[moorea21]

I have a pile of old heat sinks somewhere, and plenty of fans. I had expected to need to use some sort of forced cooling, otherwise the power capabilities of that tiny chip start to look even more insane than they actually are,

 

[PhilHornby]

Incidentally, with an inductive load, like a motor, you don't get 'in-rush' currents - you get the opposite (outrush?), when the stored energy in the coil gets released).

It's funny how sometimes I just write complete twaddle

What I said might be true of inductors in general, but not motors - when they're not spinning, they are not generating any back EMF, so the supply drives a huge current through their low resistance windings. I suppose the fact the windings are inductors helps slightly - but not enough to prevent things getting pretty hot, pretty quickly if the motor doesn't start up. Of course, once spinning it's only the difference between the supply and back EMF that is driving the current. (I learned this when I was about 12. How I managed to suddenly forget it, is a complete mystery!).

Actually, I meant it the other way round ... as in, a brushed DC motor, might just cause a Triac/SCR to drop out, when it switches. It probably depends on the size & duration of the back EMF. I think that is going to be 'appreciable' in the case of this motor!

Which is equally unuseful

In practical motors, the brushes are always in contact with the commutator. If this wasn't the case, there'd be occasions where the motor might stop in a position, from which it couldn't restart! This answer on 'stackexchange' is quite helpful.

Apparently, the varying amount of contact between brushes and commutator produce slight pulses, which can be detected to measure RPM. Something else I didn't know!