PWM Return for EDC Control

[Nick Anderson]

I want to control an EDC pump, but only in one direction.  I'm only going to use 1 of the PWM outputs, since I need to drive 6 total pumps (4 bidirectional, and 2 uni-directional).  What should I use as my return.  Have people generally just used the local ground as the return?  Should I use an output pin with a 0mA/% input to it? or should I use the sensor ground connection?

Suggestions?

Nick

[radioact]

I am driving one bi-directional pump and here is what works for me and if you implement just one side it should work for you too.

I bring three wires from the pump back to the control.
One wire I tie to the forward coil (Pin A on the MIL style connector EDC).
One wire I tie to the reverse coil (Pin D).
I tie the other two ends of each coil together for a return (Pin B & C) and run that wire back as well.

I bring the three wires back to the control and I use two PWM outputs from my controller, one for the forward wire and one for the reverse wire.
I tie the return wire to the power supply ground near the controller and not to the signal ground, I don't want to introduce any potential noise to my analogue sensors.

To operate my system, I drive a PWM signal to one of the two outputs and my pump runs proportional to the duty cycle of the output. My logic prevents it from running both PWM outputs at the same time.


For you system, just drop one of the direction wires and you should be fine.

Tyler

[jsaluk]

Are you using Pinconfig=3 or =4 for driving the EDC?

[radioact]

I am using pinconfig=4 on my code.

Tyler

[pinias]

Hello RadioAct and everyone,

I just did a project to drive the EDC forward and reverse, so I am using pin 37 and 38 of a MC050-010. This is for Danfoss 90 series pump.

I have the coils connected in series. so far it seems to work fine however there is a few detail i do not like too much.

When the machine is just started up and you try to move either direction it does not respond right away, I have to place the joystick in neutral then try again then it responds.

now, when I am moving in one direction then I go to the oposite direction some times the issue happens again, it does not responds right away, I have to move the joystick to neutral then try again then it works.

it seems to be more a mechanical issue, seems that the spool valve gets stuck but not sure yet.

my question is, have you ever seen this issue? 

if so what could be a way to fix it?

I appreciate any feedback

Thanks

[FluidPowerTom]

Pinias, I haven not ever had that issue.  I've always wired it up the way as described by RadioAct.

[pinias]

ok,

that way is connecting the coils in parallel, right?

A  input for one direction
D for opposite direction

B&C going to reference 0VDC.

the way I have them connected is in series, A for one direction, jumper between B&C and D for opposite direction, (this is a legacy connection for the company since they have always done their own circuit boards to drive the EDC)

the diffrence between the parallel and series connection will be the amount of current to be supplied.

so in terms of power, you got less power in series than in paralled (as per Danfoss S90 pump datasheet)

that is why I have tought that the back pressure on the spool valve could be the reason for the issue, however not at all sure since Danfoss says you can connected them in series but nothing else.

I hope anyone else have done this way and can give an advise

Thanks

[oiltronic]

It's not parallel because you're only driving one coil at a time.  It's what the Series-90 datasheet describes as "Single coil".

You actually get the most power (watts to the controller) when driving in the single coil configuration, as per the datasheet current limits and coil resistances.  P = I² * R.  Like so:

Coil Config	Max Current (mA)	Net resistance (ohms)	Power (mW)
Single, A-B	85	20	145
Single, C-D	85	16	116
Series	43	36	67
Parallel	85	8.9	64

The last time I had a momentarily unresponsive S90 pump there was metal debris in the controller circuit.

[pinias]

oiltronic,

thanks for clarify the misunderstood about the connections.


The way I have it connected requieres less power, but the response should not be an issue or it will be always the same in order to make sense

One of the things is that I am monitoring the output current just to be sure that I am suppling the current needed.

so far that looks ok

I will check the connections,Thanks for the advise

Regards

Antonio