Braking with radial piston motors

[hydroman247]

Hi guys, got what is probably a simple question for you.

Radial piston motors
Closed loop drive system.
Hydraulic brakes
PM is an electric motor which is on/off as you use the machine.

The issue as you can imagine is when driven up a slope and the motor is stopped, the radial piston motors do not like this. Not to worry, we have the main brakes to hold the machine steady. When we come to set off again, as the brakes release, we roll back a foot before sufficient pressure has built up to hold the motors steady again.

An expensive way around this might be pressure sensors on either side of the pumps closed loop and when we reach a certain pressure, then the brakes release but are there any other common solutions to this problem?

Thanks

 

[kcj]

How do the brakes signal to release? A & B hydrostatic loop pressures, or manually by operator releasing them?

 

[hydroman247]

Electronically and hydraulically. There is an electronic signal whenever the operator wants presses a trigger on the joystick and hydraulic pilot lines from the closed loop system through a shuttle valve operating a pilot operated spool type cartridge valve.

Initially it was just electronic and only the charge pump pressure in the system holding the machine which wasn't enough on any slope.

 

[pob786]

Hi, Why don't you introduce a sequence valve into the supply to the brake release valve, set higher than system pressure required to the hydraulic motor to "hold" it on the slope when the brake is released. By installing it on the supply to the "Brake valve", when the valve is de-energised the Brake signal is connected to tank, ensuring Brake application.
John P.

 

[hydroman247]

That was my first option however the sequence valve I have access to at the moment is none adjustable. I have already ordered an adjustable valve.

Mocked up in a simulator this does work. Sometimes I get strange oscillations but that might just be fine tuning in the simulator rather than it going to be an actual issue.

Main drive loop > shuttle valve A-B > Pressure reducing valve > POV normally dump to tank > Electronic controlled diverter normally dump to tank > Pressure relief valve > Brake release.

The reason for the PRV before the brakes is there is another method of brake disengagement which cannot be put in before the reducing valve.

 

[PersonalProfile]

I seem to recall (it was a while ago) the use of balance valves to achieve similar functionality.

Regards,
Lyle

 

[hydroman247]

Do you mean over centre valves? Counterbalance valves? This is a closed loop system and using OC valves would be really inefficient would it not?

 

[PersonalProfile]

Yes, counter balance valves.

As I mentioned it was a while ago and I did not complete the detailed engineering (I had a Subcontractor complete the detailed design and eventually this functionality requirement was dropped).

It may be worth calling your local trusted supplier and see what, if anything, they can do for you.

Regards,
Lyle

 

[hydroman247]

I am still running into problems with this braking system. Currently, I have a pilot operated valve which is operated from A&B hydrostatic loop pressure of 35 bar. Then an electronic trigger operated valve and then a pressure reducing valve to prevent the brakes seeing over 30bar.

When I drive up a slope and let go of the controls, everything is fine. I can press the trigger now but when I try and drive forwards, the 35bar needed to release the brakes is not enough to hold the machine steady, at least, not until it has rolled back a bit. I can hear the pistons slapping until enough pressure has built up again in the drive loop.

I am not sure how I can get around this as the drive pressures might only be 40-50bar on the flat but on a slope, 50bar is not enough to hold the machine steady. The slopes I am testing on are 30%+ which are pretty steep and the machine weighs in excess of 3 tons.

 

[Oldhydroman]

Your problem is like trying to do a hill start in a car – the steeper the hill (and the more heavily laden the car) the more drive effort you need to apply before you release the handbrake.

I presume your pilot operated valve (sequence valve?) is looking for a least 35 bar in the high pressure leg of the closed circuit before the brake will release. If you raise this setting for a successful hill start on the steepest slope you probably find this setting is too much when starting out on the flat.

I’ve got a nasty feeling that you are going to have to resort to some electronics. Have you considered fitting a tilt sensor (inclinometer) to the machine? The output signal of this can be used to increase the setting of a small proportional relief valve fitted to the drain line of your sequence valve. When the hill is steep the inclinometer output signal will be high, the relief valve setting will be high and the sequence valve setting will also be high – meaning the motor drive pressure has to be that much higher before the brake will release.

Actually you might need two sequence valve systems – one in each leg of the circuit. This would accommodate all four drive scenarios: driving forward uphill, reversing uphill, driving forward downhill and reversing downhill. Alternatively if you can identify the intended direction of movement you could get away with just one valve assembly and a whole bunch of logic controls.

DOL

 

[hydroman247]

I would rather not use electronics if at all possible. Cost and complexity need to be kept low.

I am looking at motion control valves on closed loop systems but it seems generally accepted that this is not a good idea. The oil in the loop gets very hot and we lose a chunk of efficiency we have been working hard to get.

Surely radial piston motors have been used on a machine before with an electric motor as the prime mover without tilt sensing and proportional brake valves.

 

[hydtools]

You need to build pump pressure to meter into the drive motor before releasing the brakes. The directional valve may need a flow blocking feature that causes the flow out of the pump to rise above drive pressure and then meter flow into the motor from that higher pressure.

Ted

 

[hydroman247]

Charge pressure is set at 15bar at the moment. Even if I hold that on for a few seconds before starting to drive, the roll back still occurs. If I try and go full speed forwards from a standing start, the roll back is worse and the pistons in the motor make an almighty noise.

When you say I need a flow blocking feature that causes the flow out of the pump to rise above drive pressure, this is not much use as drive pressure on the flat could be 50bar, on a slope it could be 180bar. The variation is the problem.

This is only really possible with proportional electronics.

 

[hydtools]

Pressure will rise and fall as the motor load changes. Motor pressure will be whatever it needs to be to drive its load when you meter flow from a higher pump pressure.

Ted

 

[hydroman247]

This is operated using a hydraulically piloted swash pump. As I push the hydraulic joystick forwards, pressure rises. I don't see how I can make the pressure rise high enough to hold the machine steady on a slope without severely affecting normal flat ground operation.

Metering flow from a higher pressure is not possible, this is a closed loop system.

 

[hydtools]

If it is a 'closed loop' system and the pump delivers directly to the motors, where does the flow go from the motors when the motors are driven backwards by the machine as it drifts backwards before you have enough flow and pressure to drive forwards? Does not 'closed loop' mean fluid flows from pump to motors back to pump? Then any increased angle of the swash pump will try to deliver flow to the motors and pressure will rise until the motors move at whatever speed the flow causes.

Ted

 

[Oldhydroman]

From what I can gather, when you park the machine on a hill and try to start off again: you release the brakes just as the pump is commanded to come out of neutral but the drive pressure hasn't yet built up enough to stop the machine running back. Your radial piston motor are themselves acting as pumps to try to pressurise that leg of the closed circuit which needs to be at the high pressure when driving up hill. The further your pump is from the motors the bigger the problem will be (actually it's all to do with the compressibility of the fluid volume between the pumps and the motors - a whole bunch of big hoses will exacerbate the problem as well).

When you are driving your car you know when to use the hill start technique - it's when you've parked on a hill. You recognise the presence of the hill and remember to do a slightly different thing to the normal start. So how about putting in a small push-button that the operator can press when doing they recognise the need to do a "hill start". The action of the button would be to interrupt the signal to the brake solenoid so the brakes didn't actually release. Then when they have held the joystick handle over far enough and long enough to feel the machine trying to drive up the hill they could let go of the button - the brakes would release and up the hill they go. A simple control, will by definition be less sophisticated than a more complicated control ... but it works on your car so it might work on this machine.

DOL

 

[hydroman247]

Yes oldhydroman, you are right. That is exactly what is happening although the hoses are not long at all, the whole loop is only 3-4m of hose, not much at all. The pump sits right near the axle.

Whilst the idea of a separate control for hill starts would probably work, it is not ideal. It is starting to look like this combination of pump and motors and control system doesn't go together very well as a mobile hydraulic drive system.

I did some pressure tests yesterday and found that when recording A&B on the pump, I could see from a starting on the slope, I would first see the charge pressure over the whole loop, I would then see "A" rise quite sharply, and "B" would lower to 0 bar temporarily as the machine rolls back and kind of reverse cavitates until the pressure builds enough for it to hold the machine.

@hydtools

Yes I understand that but a shuttle valve over A&B picks the highest pressure and it only needs 35 bar to begin opening the brakes. If I had a PLC I could put pressure sensors and feedback response on the joystick to work out what is going on and pick the appropriate action but this is almost entirely hydraulic.

 

[Oldhydroman]

Let’s imagine you are trying to make a very gentle start in the forwards direction having previously parked on a hill (facing up hill). You might only push the joystick over a short way so you only get a tiny flow from the pump (the fact that it is an electric motor drive is irrelevant). The pressure starts to build up in the driving leg of the circuit (let’s call that A) and you only have boost pressure in the other leg (called B obviously).

As soon as the A pressure reaches 35 bar that’s enough to release the brakes – but not enough to hold the vehicle on the slope. The radial piston motors run backwards (acting as pumps) and they take the oil out of the B side and push it into the A side. To recap: just at this instant in time the pump is pushing oil into A (and sucking out of B) and the motors are pushing oil into A (and sucking out of B). The only thing pushing oil into B will be your boost pump – and if the B pressure is dropping to zero then the boost pump is not keeping up with the flow demand.

So your boost pressure is dropping to zero –the pump won’t like that one little bit. The pump will start to cavitate and the flow you were expecting it to deliver isn’t going to be there. This might also be contributing to the noise. The motors (running backwards and acting like pumps) will also be cavitating – and they won’t like that either (more noise).

Here’s a funny thing to think on as well… if the motors do more than one revolution while they are running backwards (eccentric crank type motors – Staffa, Calzoni etc.) or if the pistons in the motor do more than one stroke (external cam type motors – Poclain, Hagglunds etc.) then the cavitation means that the piston chambers are not being refilled properly on their “suction” strokes. If the chambers are not completely full then the pistons provide no resistance to the rotation of the shaft when they are on their “delivery” strokes. The motors freewheel [noisily] and this will exacerbate the roll back.

You might want to try fitting a small accumulator to the boost pump outlet (and a check valve to stop the accumulator back driving through the boost pump). The accumulator will help to support your boost pressure during the transient conditions of a “hill start”.

DOL

 

[PersonalProfile]

If the machine does not need to be reversed, maybe consider a mechanical anti run back device.

Regards,
Lyle