Hydraulic Proportional Valve with no amp?

JRoss

Lifetime Supporting Member
Join Date
Sep 2006
Location
Dillsburg, PA
Posts
184
Has anyone on hear ever heard of controlling a proportional hydraulic valve directly from a PLC with a PWM output and SSR, instead of using an analog output to an amplifier? I have an OEM customer that wants to do this to save money on their machines.
 
As in my algorithm would continue to provide a signal, or that the valve would require one to stay fully closed?
 
As I understand it, from a tech point of view, the dither provides for continuous very small movements to the valve to lessen the chances of the valve momentarily sticking when given another movement signal. But it's been a while. Check on the valve's requirements.
 
Thanks for the feedback! I'm wondering if you could give more specific reasons why not? Honestly my reaction when they ask me was the same but I don't have the experience with hydraulic equipment to back it up. The controls engineer there is on my side but the project manager seems really set on trying to save the cost of the amplifiers.
 
Thanks for the feedback! I'm wondering if you could give more specific reasons why not? Honestly my reaction when they ask me was the same but I don't have the experience with hydraulic equipment to back it up. The controls engineer there is on my side but the project manager seems really set on trying to save the cost of the amplifiers.
How fast do you think you can PWM and get the SSR to switch?
It probably will not be fast enough. That is the PWM frequency will not be fast enough. Next, the PLC will probably not be fast enough if your are trying to do closed loop control.
What are you trying to do?
 
I do not think the the analog output can supply the power to hydraulic valve that the amplifier can.

james
I doubt it too.
Most hydraulic amplifiers use PWM to move the spool.
I was assuming JRoss's intentions were to use a PWM signal to an SSR that would control the current to the valve.
The output from the PLC would not be analog.
However, I think there are two problems with the SSR.
1. I doubt it could switch on and off fast enough
2. The SSR could only control positive or negative current.

Small DC motors use an H-bridge.
https://en.wikipedia.org/wiki/H_bridge
There are H-Bridge chips.
This would allow positive and negative current.

Bosch uses a spring to force the spool to one side to a safety position. I think the solenoid only need a positive current to force the spool to the neutral position.
This is safe and simple but I have always wondered if the response is the same shifting the spool in both directions. So far I have not anything to indicate the response is not the same.

We need to find out what kind of valve is going to be used. Provide a link to the pdf.
 
I don't know the exact valve they are using.
This is a redesign of a previous machine. In that one, we're using it for simple position control. There's a linear encoder giving feedback to the PLC. I have basic motion control being handled in the PLC, calculating a speed curve. I then have a simple PI loop that controls an analog output to a proportional valve with built in electronics to match the speed command. It's not as precise as servo control, but it gets the job done to their satisfaction.
Now they are working to reduce cost, hence this question.
 
You should read my articles
https://www.hydraulicspneumatics.com/controls-instrumentation/plc-vs-motion-controllers
and
https://www.hydraulicspneumatics.co...tion/hydraulic-servo-control-using-plc-part-2

You may be OK doing it from a PLC if the PLC was using a +/- 10 volt going to a valve with on board electronics that will accept the +/- 10 volts.

I still cringe. Just the fact that you don't know the facts is a problem. At least you have a better idea of what questions to ask.

Most PLCs do not have the right kind of PID. See the last part of the second article about using a PID with both the proportional and derivative gain act on changes in the actual position only. Most PIDs do not have an option to have the proportional gain act on the changes in position.

The problem with the method is that the response will be slow but the motion will be smooth.
It would be like having 4 low pass filters acting on the error between the target position and the actual position.
 
For super down and dirty application it may work but I would think the performance will be worse than the valve with on-board electronics.

True, the on-board electronics are likely driving the coil with PWM but there is also probably some dither added which won't be in the standard PLC PWM output block (maybe some special command has it?) and maybe other nuances that the valve mfg. put in there that a plain PWM block won't have.

Are you doing true position control or just ramping speeds and running into hard stops?
 
One of the biggest reasons to use the amplifier as well will be future calibrations. You know the 4-20ma from the A/O isn't going to shift but with age, leakage and wear the valve will shift. So how are you going to compensate and calibrate the valve from the PLC PWM/O (that hasn't shifted)? More programming = more money in a different place. And how many instrument techs have the instruments, PLC software and are used to calibrating PWM driven outputs from a PLC? More chance of error and problems.

Again, I fall back to best practices and expected design & operation is always the best design. No offense but sounds like the project manager knows just enough to be dangerous. Don't build it cheap, build it right. If you created a hazard evaluation during your design phase, it would most likely show that safety and long term maintenance cost will outrank this imagined cost saving on build. If you don't have a hazard evaluation, ask your PM why not?
 
I'm not sure I was clear. I did the programming on the original machine, and it is working well. The valves accept +/- 10V from a PLC analog output. We're tracking position/speed of the axes with encoders. I programmed very simplistic motion control by calculating a speed curve of the axis given accel, current position, and target position. This gives me command speed at any given point. Then I use a simple PID algorithm to control the analog output to match that speed. Once the actual position is within a set range of the command position, I stop motion. I also calculate a command position, but only use that for a position error alarm that stops motion.

There are 8 axes that need to start and stop at the same time, and I'm able to do that repeatably with the above approach. I think that it's at least in part due to a well-designed hydraulic system that stops and holds position without drifting.

Elegant? No. Precise? Enough for the application. The customer was very happy with the result. Well, except for the cost. They need to reduce the cost of the machine to make it marketable. I'm trying to keep them from gutting out so much that the performance suffers.
 

Similar Topics

We have a proportional valve where we suspect that we may be at the performance limit. With the performance limit, I mean what I see in a curve of...
Replies
2
Views
2,199
Hello! I have a signal converter, reference PEM - 020 - G24 , which converts a signal of 0-10 V to a current signal that goes to a proportional...
Replies
1
Views
1,868
In the attached picture, you can see a hydraulic schematic of our installation with a proportional valve (Atos) with integrated force control...
Replies
26
Views
11,303
Hi, I am an inventor looking for a way to automate a block-making press I made out of a log splitter. It makes blocks for building homes from...
Replies
11
Views
7,175
Hello guys, The processor is a SLC 5/04 using RSlogix 500. Have a question about a proportional valve on our hydraulic system on one of our...
Replies
3
Views
3,586
Back
Top Bottom