Gravity and lowering hydraulics 4

[hydroman247]

Hi everyone.

I have a question for you. Do any big manufacturers of machinery use gravity to lower hydraulic booms? Obviously most are powered down using counterbalance valves which uses energy and in this day and age, efficiency is more important than ever.

I can't find anything online but have been told that JCB have some sort of gravity assisted system.
Is there anything else I should know about these systems?

Thanks

 

[hydtools]

Even if gravity assisted, all loads coming down are gravity assisted, some means must be provided to open a holding valve to permit release of fluid from the lifting cylinder whether powered down or gravity assisted.

Ted

 

[hydroman247]

Apart from the obvious gravity assistance, the pump is still having to create flow and the various valves are producing pressure which reduce efficiency.

Are there any valves which will use a much smaller amount of power but still provide a controlled descent?

 

[hydtools]

Electric valves

And it is lost work that reduces efficiency. Controlling the decent by controlling pressure results in lost work. Work that was done to raise the load is being wasted as the load is lowered. Unless you see control system work as lost work.

Ted

 

[hydroman247]

Electric valves. Hmmm. We use fully proportional electronic valves in other places but for lowering loads, they become unstable unless we use a counterbalance valve which negates the efficiency of just electronic control.

Does anyone know of an electronic valve system without counter balances that can be used to smoothly lower loads.

Thanks for all the help.

 

[Oldhydroman]

I'm interested in the nature of the instability you have found when using proportional valves to lower a load. Can you tell us more? You might be discounting them unnecessarily.

There are lots of hydraulic machines that use gravity alone to lower a load: vehicle tail lifts, tipper trucks, fork lift trucks, hydraulic elevators (think shopping mall), engine hoists, trolley jacks etc. The trick is to identify exactly what you are looking for in the system and choose suitable components and control strategies.

Here’s a few pointers:
• Presumably you want the circuit to be leak-free – or can you tolerate the lifted load creeping down over a matter of several minutes/hours?
• Do you want the load to come down at some nominally fixed speed, an absolutely constant speed or do you want to be able to vary the speed as you go?
• Do you need any acceleration/deceleration control or is simple “lower”/ “stop” sufficient?
• Are there massive variations in the working temperature of the fluid?

If you have a reasonably fixed load, a fairly consistent operating temperature and you want to lower at a nominally fixed rate, then a simple orifice and a solenoid operated poppet valve (leak free) will suffice. Or the poppet valve could be manually operated (via a knob, lever, handle, linkage or cable). Or both … it is common on scissor lifts for the normal lowering function to be solenoid operated and the emergency lowering function (from ground level) to be operated by a cable release attached to the manual override on the solenoid valve. The beauty of the orifice approach is that it is simple and relatively insensitive to small changes in load (the speed will go up in proportion to the square root of the load: the load has to quadruple for the speed to double).

If you want a [sort of] fixed speed with varying fluid temperatures then make sure that there is little pressure drop in your pipework and valve (be generous with the sizes) and that your orifice is the “sharp edge” type.

If you want a fixed speed under significantly varying loads then you will need a type of pressure compensated flow control valve. Some of these are specially constructed for load lowering – look for SB and SQ valves from Hawe as an example of this.

http://www.hawe-hydraulik.ru/pages/hawe/pdf/D6920-en.pdf

Be careful when using an ordinary pressure compensated flow control valve in a lowering circuit. Explanation: when you close your solenoid valve the flow control valve’s hydrostat will open fully. Then when you next open the solenoid valve the hydrostat has to move back to the “correct” position from the starting point of “wide open”. In the few milliseconds it takes to reach the correct position the hydrostat will be too far open – the flow will initially be too high and then re-establish the right value. You will feel this as a “jump” when you start to lower, it might be of no significance but it makes it difficult to set the position accurately if you ever have to stop before reaching the bottom. It is possible to arrange for external closing of the hydrostat to remove the jump but it is hard to incorporate such a technique without spoiling the leak-free nature of the circuit.

If you want variable control of the lowering speed from some remote position then an electrically adjustable proportional valve combined with a solenoid operated poppet valve might be the way. (The poppet valve is only needed if you want the circuit to be leak free because a proportional throttle valve is unlikely to be so.)

There are lots of proportional valves that will do what you want. Some have built in pressure compensation (check out Wandfluh) and some, if force controlled (non-feedback type) rather than stroke controlled (feedback type), will allow the flow forces to modulate the opening of the throttle and this gives a rough pressure compensation effect. I’m wondering now if your previous [bad] experience of proportional valves in load lowering circuits was because you were using a stroke controlled valve with a very high pressure differential.

DOL

 

[hydroman247]

Thanks for the brilliant response. I will do some more research over the weekend. The application is for cherry pickers so no leakage and safety are top priority followed by proportional control which is what makes it more awkward.

I will type out a proper reply once I have had a chance to have a look at some of your suggestions.

Thanks!

 

[hydroman247]

Just also to point out the flow rates I am working with are 0-50l/min and the systems are mobile.

 

[hydtools]

What is your target efficiency gain? What is your current system efficiency?

Ted

 

[IceStationZebra]

"The application is for cherry pickers so no leakage and safety are top priority followed by proportional control which is what makes it more awkward."

This application requires that you use counterbalance or PO-check valves in case of a hose break, so gravity alone will never be an option.

ISZ

 

[Oldhydroman]

A very good point ISZ.

But this is a global forum and hydroman247 hasn't said which part of the world he is in. Specific regulations about coping with hose bursts will most likely be on a national or federal basis - even though "best practice" dictates that some sort of protection will definitely be required.

This will be a man-riding machine so provision has to be made for someone on the ground to be able to lower the boom in the event of the operator in the basket becoming incapacitated (electrocuted from overhead lines that he didn't notice on the way up, knocked himself out by lifting the basket without checking that there wasn't a beam right above his head - all the usual scenarios). The boom also has to be capable of being lowered to the ground in the event of the hose burst valve closing because of... a burst hose. You also have to consider how to get back down in the event of no pressure being available to pilot open the CB valve because the engine has "bust a gut", or there has been a fire in the battery compartment, or the pump has "blown a gasket", or the prime mover - whatever it is - has plain and simple run out of juice. So, in all events, there has to be some way of circumventing any pilot operated valve which has been incorporated.

If the load lowering valves, whatever type they may be, are installed directly onto the port of the boom cylinder then this could probably satisfy any practical or legislative requirement to guard against hose failure.

DOL

 

[hydroman247]

I am based in the UK. Yes counterbalances have to be used which is why this is a bit more awkward without custom made cartridges. The systems I have heard about have all been non man-riding so this might not be an option.

I have not had a chance to do any more research over the past couple of days but I will make a proper reply in a bit.

 

[Oldhydroman]

I would strongly suggest you check the exact wording of the applicable legislation and ISO standards. For sure it will say something generic about provision to guard against uncontrolled and excessive lowering speeds in the event of a burst hose - but this does not automatically mean "counterbalance" valve. And, as a lot of the hydraulic engineers reading this will agree, all sorts of valves have the title "counterbalance valve", and each of those valves also has several different titles depending on what the marketing boys deigned to call it in the catalog: overcenter, load holding, load lowering, motion control, braking, counterbalance with pilot assist, boom lock, the list goes on and on.

We used to have a man-riding scissor lift at one place I worked at in the UK. It was battery operated and the motor driven gear pump ran at a variable rpm to control the speed of the lift but lowering was gravity only, at a fixed speed and activated by a solenoid check valve mounted directly on the port of the single acting lift cylinder. This circuit had all the necessary approvals and certification - it did not have a "counterbalance" valve (something piloted open by a hydraulic pressure signal). I don't think "energy efficiency" per se was the object of the exercise - well it was but only to extract maximum endurance from a single charge of the batteries: the pump didn't have to run when lowering and that's really worth something when trying to get a full day's work out of an overnight charge.

I also worked on some vehicle tail lifts, which are definitely man-riding, and these had two solenoid check valves in series - both solenoids had to be energized to lower the load, so there's some degree of redundancy built in, i.e., both valves had to fail before the load could drop. The valves were [deliberately] different styles of construction and from different manufacturers. And there was a load lowering flow control valve (one of those Hawe inserts) - and the drop was only 3 foot (I suppose the driver/crew could jump off if they didn't like what was happening).

Now I didn't tell you this right, but have a quick peek at some of the circuit diagrams for their splendid range of powered access equipment and mobile elevated work platforms that Genie helpfully provide on their website. This is just for academic interest in information already in the public domain you understand - there's no parts lists and no design data anyway so you can't copy what they've done.

http://www.genielift.co.uk/en/sales-and-support/manuals/index.htm

and select the link to schematics and diagrams.

In particular look at:

http://www.gogenielift.com/service/schematics/Z_Boom/Z45-22_DC.pdf

and peruse page 33 (boom reach lift and boom elevate circuits)

Look round the schematics of a few of the other machines and you will see all sorts of hydraulic circuits. Yes, some use double acting cylinders with counterbalance valves (some single and some dual) but you will also see circuits showing single acting cylinders using solenoid check valves with in-line orifices controlling the lowering of the booms etc. You will see some of the solenoid check valves with dual coils (not opposing each other like in a three position valve but assisting each other or acting as duty/standby or main/reserve). You will also see throttle valves in parallel with the solenoid valves (normally closed but opened for controlled manual release) and you will see solenoid check valves with manual overrides (some operated by cable release).

Hope this helps.

DOL

 

[hydroman247]

Thanks for all that info.

I do need to do more research on the regulations here in the UK. As mentioned above, there needs to be hose burst protection and a way of getting the operator down with no power from the base.

Most current systems used counterbalance valves because they are so smooth and by counterbalance, I mean actual counterbalances. PO checks are not smooth at all.

Just rolling thoughts around in my head here but I am thinking it won't be possible to have full electro-proportional control because of the safety issues, but perhaps when operated from the cage to full speed it opens up a bypass circuit with a proportional flow control valve and orifice combo once the operator goes faster than a set speed. As flow goes through this, pressure would drop at the counterbalance and that would close smoothly. Not sure how this would go in practice.

I have to do some more thinking.

 

[hydromech]

I have watched this post from afar, but I can no longer resist...

UK regulations are somewhat sketchy, but common sense would be to put velocity check valves on any hoses. A risk assessment should show that velocity check valves are not necessary on rigid pipework.

Lowering booms under purely gravity acceleration is difficult because of the different viscosity of the fluid depending on the ambient air temperature. The rate of decent changes with increase in temperature and it's not a linear relationship. Likewise, tuning a proportional / PID controller to suit all scenarios will be troublesome and expensive.

Unless you are concerned about energy recovery, it is not worth considering the efficiency of lowering.

Counterbalance valves are set to nominally 10% above the maximum working pressure in the cylinder. A counterbalance valve is a relief valve. If you add a pilot signal to it to open it, it becomes an over-centre valve. In either valve, if the poppet get stuck open, the load will drop. In these circumstances, the flow rate through the valve is still limited by the open area and the boom will not have an uncontrolled decent.

As part of the CE marking process, you are expected to cater for reasonable failure modes only. If you make provision for reasonable failure modes, then you covered from a legal point of view...as long as you provide technical data / operation instructions to high the dangers associated with using the machine.

Scissor lifts, cheery pickers, boom lifts, cranes...whatever. The all rely on motion control valves mounted directly to the cylinder to support the load and provide controlled decent. Above and beyond that, if you want decent under gravity, you will need to fit a pressure compensated flow control valve in a separate line back to the reservoir.

If you have ever seen the James bond film "Casino Royale", you will remember the scene where Mr Bond jumps on to a scissor lift, smacks the control valve with a spanner, then rides the lift from top to bottom in 2 seconds before jumping off at the bottom. Can you imagine that happening in real life?

Adrian

 

[hydroman247]

Just a velocity check valve on the cylinders and then electro-proportional valves would be the ideal option but like you say, would need a lot of tuning to get it all nice.

Efficiency is what I am aiming for and currently using a lot of power to descend when gravity is all around us seems like a waste.

Do you see anyway around the viscosity change problem? These machines could be working in -20c or 45c and be running all day. The only way I can think of is electro-proportional. The flow rates will still change but the operator would be able to slow it down. Having some sort of temperature sensor and adjusting the solenoid control system is a bit OTT and unnecessary.

A separate line back to base is not really an option. Minimising weight is crucial and adding metres of hose is not ideal.

As for James Bond, I think everyone who works with hydraulics, cringes at that moment.

 

[hydromech]

Indeed...

The only way to negate viscosity issues is to use aviation type synthetic fluid such as Skydrol. They have a good viscosity index and remain relatively stable across the temperature range. That said, there is still a change in viscosity that can scupper your plans on a hot day and sourcing valves with ethylene propylene seals is a nightmare.

Can you fit a small hand pump in the cradle that the operator can use to open the motion control valve...a small 6mm pipe from the cradle to the cylinder?

Or what about a 12V pump unit and a proportional pressure reducer with a potentiometer. As the pot is turned, the pilot pressure to the motion control valve increases. The motion control valve opens, the cradle lowers under gravity. If you back the pot off, the motion control closes and the cradle stops.

Power requirement would be a few hundred watts at most...hardly inefficient.

Lots of options...

Adrian

 

[hydtools]

hydroman, how much power is used to descend? A pilot signal does not represent much power. Controlled descent still requires pressure and/or flow control of the fluid exiting the cylinder. Where are the efficiency gains?

Ted

 

[hydroman247]

I don't think Skydrol will be an option. 1. It is horrible stuff and 2. It is expensive.

These machines can be quite big and weight in the cage is kept at an absolute minimum, as is the weight of the booms etc. The only things going up to the cage are control system wires.

The Pot idea is good but I don't think we can have a pot that operates an actuator directly and stays moving once let go for safety reasons. We already have a pot in the design which controls the spools in the base valve blocks.

Power usage off the batteries can use over 150amps at the moment.

 

[Oldhydroman]

What you're trying to do here (bring a load down safely and under gravity alone) isn't THAT difficult. Problems do start to appear, however, if you require the speed to be absolutely consistent regardless of changes in load, machine geometry and temperature. Can you live with some small variation in speed? Would the world end if you had to rely on the [trained/responsible] operator to have some input to the process? Remember, the simpler you keep your circuit the less there will be to go wrong and the more straightforward will be the risk analysis you have to do to show that it is safe under all reasonably forseeable failure modes.

Lots of systems have operator controlled "hand-to-eye" coordination to "automatically" provide changes of control input. Humans are particularly good at doing this to acheive the desired velocity profile in response to variations in load. You do it yourself whenever you apply the brakes on your car - giving more foot effort if you don't seem to be slowing down enough because you temporarily forgot that you've just loaded three dozen paving slabs into the trunk (company car obviously).

I've just been watching the tennis matches at Wimbledon in the UK - how can a player work out if the approaching ball is going to land outside the line? They actually solve a quadratic equation in their head ...but when put like that it sounds more fabulous than the reality of the situation. What's my point? I don't think you need to take absolutely everything out of the hands of the operator, they can be a vital part in your control strategy. Ask youself this: how is the maximum descent speed controlled when people are working from a man basket hanging off a crane hook?

So, on your cherry picker descent circuit you could have, built into a manifold mounted directly onto your cylinder, a series connection of several valves - each with a specific task:

1) A fixed setting, pressure compensated, load lowering valve insert [or a simple sharp edged orifice]. This would set the maximum lowering speed that would occur in the event of any rupture or failure of a downstream component. Maximum speed would be determined at maximum load and highest oil temperature, i.e. maximum pressure and minimum viscosity. Alternatively the first valve could be a hose break valve which slammed shut if the fluid velocity ever went too high. Which would be the preferable scenario in the event of a massive failure of the downstream hydraulic system: come down at a [safe] maximum speed (probably too slow for 007), or stay up there until it's fixed (007 wouldn't like that one bit)?

2) A solenoid operated check valve with a leak free construction - this valve provides the secure load holding with the ability to switch open when required. There's lots of precidents for the use of this valve. And very close to this...

3) A normally closed (but innevitably not leak free) electro-hydraulic throttle valve. The solenoid would be driven by a proportional valve amplifier - the input to which could come from a joystick in the hands of the operator. This valve would control the instantaneous lowering speed, a speed which could be varied steplessly between: very slight creep [no effective proportional signal yet but with the solenoid check valve open] and, full speed [a value determined by: the size of the valve, the maximum solenoid current you allow, the maximum pressure difference and the minimum viscosity]. You could make this electro-hydraulic throttle valve pressure compensated as well if you wanted to. It's hard (but not impossible) to find a sharp edge version of the electro-hydraulic throttle valve so the typical valve won't be temperature compensated.

Within your electrical control system you would need to co-ordinate the energisation of the check valve solenoid and the start of control of the electro-hydraulic throttle valve. Some proportional valve amplifiers can provide this function with an auxilliary switched output, or you could use some sort of trip amplifier to monitor the signal from the operator's joystick. Additional security could come from the use of a pushbutton switch in the end of the joystick. Or you could have a cam operated switch which energised the solenoid as soon as the joystick moved out of neutral. Combine this with a mechanical safety gate arrangement to guard against inadvertently knocking the joystick out of neutral. This coordination of solenoid energisation and start of control is no more difficult than sequencing the release of a hydraulic motor brake with the energisation of the DCV that drives that motor.

You still need the manual release system (not discussed above) and I agree with Adrian, to guard against common-mode failures this really should have its own line back to tank. But it needn't be a big line and it would only need to have a high pressure capability on the upstream side of whatever device you use as your manual release valve.

You can mitigate (but not remove) the effects of extremes of temperature by using an oil with an exceptionaly high viscosity index. [The viscosity index of an oil is a measure of the degree to which the viscosity varies with temperature: the higher the index the less the viscosity changes.] Ordinary HM oils have a viscosity index of around 100. Not-very-difficult-to-find HV oils will have a viscosity index of about 150. If you look a bit harder you will be able to find oils with a viscosity index in the 300-450 region. Try "Mobil Univis HVI" or "Shell APR HC 13". Some aviation hydraulic fluids have suitable viscosity ranges but, unless you move in those circles, they can be awkward to source; try OM-18 (Aeroshell fluid 4).

It wouldn't be too difficult to give the operator a little "help" to compensate for changes in fluid viscosity without going OTT. Suppose that the supply to the joystick potentiometer came from a PLC analogue output rather than the valve amplifier itself. Then suppose the PLC had an analogue input of fluid temperature (this should be the temperature in/of the cylinder). Then in the PLC programme you could apply a simple linear relationship between the magnitude of the voltage output to the joystick and the extent of chilling of the oil. When the oil was very, very cold the joystick supply voltage could be a little higher and the proportional valve would be allowed to open a little more. I'm not pretending that it's a full blown temperature compensation loop - it's just continued reliance on the operators' hand-to-eye co-ordination abilites but giving them a little bit of help so they don't have to compensate quite so much themselves.

With such a scheme you can then modify the voltage output algorithm so that the joystick supply voltage reduces to a small value as the basket approaches some "nearly-fully-down" position (detected by some failsafe limit switch arrangement).

Others may scoff at this simplistic approach to motion control and temperature "compensation", suggesting instead that you should use full blown, software driven, extensively proven and fully configurable, multi-axis capability computerised closed loop controllers. These controllers are absolutely fantastic. But do you ever wonder how our predecessors managed to get anything to work at all without these kinds of devices available to them? To quote Maslow:

"I suppose it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail."

So, my remainig tips would be: make sure you are not unnecessarily discounting simple and robust solutions. Allow the operator to be human and control his/her own destiny. Make it so the worst thing that can happen is a minor jolt (no worse than driving over a pot-hole) if they inadvertently land the cherry picker at its maximum speed.

DOL