Electro-hydraulic actuator

[Steve_88]

I need to control an electro hydraulic actuator from a PIC. I've some experience with doing this, but the actuator schematic has me a bit confused. The data sheet indicates that you swap the + and - leads to change direction. An h bridge should work, but the data sheet shows doing it with a SPDT switch/relay. I've attached the data sheet, can someone clue me in here?
Thanks

[HenrikOlsson]

Hi,
It seems to me like that SPDT (with center position off) switch controlls two relays/contactors (CR1 and CR2). The contacts of these relays/contactors forms a H-bridge.

/Henrik.

[mackrackit]

Their schematic is confusing..
You will need two switches/relays.
Page three of the attachment figure B should help.

[Steve_88]

Thanks for taking a look at it. I assumed it was an hbridge switching polarity with a secondary relay to cut power. The unit will draw upwards of 40 amps max, with a typical draw of around 20 amps at 12 vdc.
I've started looking for suitable relays, but have not found much other than automotive type relays. I have not found anything with a DPDT configuration to change the polarity with. I did find a few SPDT units but still looking. Any suggestions would be great!
Thanks, have a great Holidays

[NickMu]

Replace the SPDT unit with two SPNO units.
Your logic should actuate none or only one of the SPNO units at any given time.
Both SPNO units ON is what you want to stay away from.

Nick

[mackrackit]

I always seem to end up up using automotive SPDT for things like this, mainly because they are easy to find and reliable.
Use two and have the coils wired together so they are both energized at the same time giving you a DPDT. Be sure to use a fuse just in case....

[HenrikOlsson]

Hi,
I think you are making it more complicated than it really is....
If all you need is to be able to drive the motor in both directions as well as not drive it at all you need two SPDT relays (RE1 and RE2 in the below (very crude) schematic), each driven by an output on your PIC.

Here's the crude picture:



Both relays are, as usual, drawn in their "inactive" state, both sides of the motor is connected to the positive side of the supply and no current will flow thru the motor - it won't go anywhere. To run the motor in one direction activate one of the relays, to run it in the other activate the other relay. If you activate both relays both poles of the motor will be connected to the negative side of the supply and the motor will not run.

/Henrik.

[amgen]

yes, and a good ckt for spike reduction and inductive energy dampening for bi-directional ckt, using 1 full wave bridge.

[Steve_88]

Thanks for the explanations/examples, that is a huge help
I'm not completely clear on using the full wave bridge though, are you using just two of the four diodes?
Thanks again

[Chirpy]

the 4 diodes in full wave bridge will convert the positive and negative parts of AC waveform into a positive only waveform that has both of them bouncing on the positive side. to convert it to a clean dc signal, you need a large capacitor which will take care of all of the ripples and will create a clean dc signal. if you only used two diodes (half wave), it just cuts the ac signal's negative off and only gives out the positive pulses from the ac. if you look at the waveform through an oscilloscope of a half wave bridge, where the negative pulses were at, it's just ground/0v.

easier way to think of it is full wave bridge's waveform is "bump bump bump etc" while half wave is "bump nothing bump nothing etc".

[HenrikOlsson]

Chirpy,
That's what a full wave rectifier normally is used for but it has nothing to do with this application.

Here it is being used to provide a circulating path for the motor current when the contact on either side switches from one state to the other.
* Lets say the left contact is in the "up position" and the right contact in the "down position".
* The current is flowing from left to right thru the motor.
* Now the left hand side contact is switched to its down position in order to stop the motor.
* Just as the contact opens the inductance of the winding in the motor will cause the voltage on the "left side" to rise abrubtly, if the diodes wasn't there this voltage would casue an arc to develop across the now open left contact. Because of the diodes the voltage spike is clamped because there's a path thru the motor, down thru the right hand contact and back up thru the left hand side diode.

The reason for using a bridge rectifier like that is, I guess, due the spade terminals making it easy to connect when using an automotive type relay like the one in the picture. Two single diodes would work just as fine (and be cheaper).

/Henrik.

[amgen]

H, great explanation for the free-wheeling diodes circuit.
about the bridge...... my experience but debatable
-if the motor is 20-30-40 amps, the diodes should be at least 15 or 20 amp..... easy hookup with bridge (like you said)
- 1,2, 6 or 10 amp diodes could short out..... too small ( even though the current is of short duration)
-I saw 1 implementation where they put the '+' bridge connection to the '+' dc, but I think that those diodes are not needed and don't ever
carry any current.
- also should try to use FAST SWITCHING DIODES/BRIDGE, a little more expensive, reduces the doide turn-on-time, lowering voltage spike further.

don

[HenrikOlsson]

Hi Don,
Actually, lets say the situation is the same as before, current flowing from left to right thru the motor. This time, in order to stop the motor, you switch the right hand side contact to its upper position. While the right hand side contact is open can you show me path for the current to circulate?

If you connect a diode from each side of the motor to the positive rail as well there will be a path.

The difference between a normal diode and a "fast switching diode" is not the turn-on-time, it's the turn-off-time or reverese recovery time as it's called.
When the diode is forward biased and conducting current from anode to cathode its PN junction is charged. When the voltage across the diode is reversed this charge needs to be depleted before the diode is actually blocking the current. This means that the diode is actually conducting current backwards for a (very) short time before it starts to block it.

/Henrik.

[Steve_88]

I have a few 1500 watt, 20 + vdc unidirectional transient voltage suppressors around. Sounds like these would do nicely?
Also, I'm having little luck finding suitable relays. I was hoping to stay away from the large can types. It seems most automotive types listed at Newark, Mouser, Digikey top out 40 to 50 amps.
I was hoping for a bit of a buffer, the actuator tops out at 50 maps. For my application I'm at the low end of the pressure range, it will be typically drawing around 20 amps.
Thanks

[Normnet]

I have a few 1500 watt, 20 + vdc unidirectional transient voltage suppressors around. Sounds like these would do nicely?
Also, I'm having little luck finding suitable relays. I was hoping to stay away from the large can types. It seems most automotive types listed at Newark, Mouser, Digikey top out 40 to 50 amps.
I was hoping for a bit of a buffer, the actuator tops out at 50 maps. For my application I'm at the low end of the pressure range, it will be typically drawing around 20 amps.
Thanks

See Grainger for contactors.


Norm

[wdmagic]

ive done this using a home made MOSFET H-Bridge, using a fairchild FDP8440 It takes 4(H) and 4 diodes but ive handled a DC motor 48V and 80amps using this circuit

[HenrikOlsson]

Yes, although conciderably more complicated and easier to destroy than two SPDT relays a MOSFET bridge IS an option. But it takes a bit more than 4 MOSFETs - you need proper driving circuitry for them and most high side drivers can't keep their output on constantly because they need to refresh their bootstrap capacitor. So you really need to understand the operation of the bridge and design it with the correct driver circuitry if the high side needs to be at 100% dutycycle.

Another option is 4 solid state relays, like this one on EBAY? It's rated for 50A DC and it as simple to use as driving a LED. You need to be very careful not to turn on both relays in the same leg at the same time though and you probably need to introdce some deadtime to prevent shoot-thru. That's why two SPDT relays are so handy for this - they can not short out the supply. If all you need is to drive the actuator forward and reverse I really think two 50A SPDT automotive type relays and 4 diodes is your best and cheapest bet.

/Henrik.

[amgen]

check these reversing contactors, here in the states.......
http://www.surpluscenter.com/item.as...tname=electric
they seem to have a lot of s-it
don

[Steve_88]

I'm tempted to try out an h bridge for this motor. The duty cycle will be near 50/50 at various loads and will need to switch 1,000 + times a day a couple of weeks in a year. I found this link, looks perfect for my application. http://www.hvlabs.com/hbridge.html. What do you all think?
Thanks