Hydraulic Pressure Data Logging 2

[HydroBuff]

Hi folks,
I am having a large amount of hydraulic pumps fail on a certain brand and model of equipment so I want to add a device to log and monitor pressure.
The machine has absolutely no HMI software or plc controls, so I am wondering if anyone would have any recommendations on some type of inexpensive pressure transducer that I could install on a a pressure line that would log data every time the machine spikes above a preset amount. Then later I could come back and view results. It would need to log for several weeks, or months at a time.

If any of you can suggest a device that meets the criteria above, I would appreciate it.
Thanks,

 

[3DDave]

What is the pressure range?
What is the budget?
What is driving the pumps?
What is attached to the pumps?
Is the attachment via hardline or by hose?

 

[BEMPE16524]

you can write simple software using VB.Net (I think the most easiest programming language) and you can create simple HMI within the Visual Studio. then you can communicate with the pressure transducer using RS232, 485, etc. in your program, you can include logging of your data at particular time interval.
It is basically very simple.
you may need to use some simple microcontroller such as Arduino, Rasberry, or even a MicroBit can do the job to communicate and read the signal from the transducer and convert it to the digital input of your computer for the HMI.

as for the pressure transducer, there are a lot now you can buy from online store.

if you are looking for a transducer that can do data logging at same time, I am not aware of that.

 

[HydroBuff]

What is the pressure range?
What is the budget?
What is driving the pumps?
What is attached to the pumps?
Is the attachment via hardline or by hose?

I'm sorry, like usual it looks like I didn't post enough of the pertinent details.
The main pressure relief is set at 3600 psi., but I expect that it may be spiking to more like 6000psi going by the types of failures we have had.
The budget? I hadn't thought of that yet.Actually probably under $500.
The pumps are Parker tripple gear pumps on a pto from any random triaxle truck
The pump pressure is being used by Parker F130 valves to control various booms and other functions.
All the attachements are made by hose. I already have diagnostic quick connects in place. I can adapt or whatever to make any type of connection work.

Best Regards,

 

[HydroBuff]

you can write simple software using VB.Net (I think the most easiest programming language) and you can create simple HMI within the Visual Studio. then you can communicate with the pressure transducer using RS232, 485, etc. in your program, you can include logging of your data at particular time interval.
It is basically very simple.
you may need to use some simple microcontroller such as Arduino, Rasberry, or even a MicroBit can do the job to communicate and read the signal from the transducer and convert it to the digital input of your computer for the HMI.

as for the pressure transducer, there are a lot now you can buy from online store.

if you are looking for a transducer that can do data logging at same time, I am not aware of that.

I was really looking for some simple off the shelf software, but I would be open to programming if neccesary.

 

[PNachtwey]

The budget? I hadn't thought of that yet.Actually probably under $500.

Best Regards,

You can get a HMI for that price. I doubt you will get a sensor and HMI for that price. The HMI will not have any data logging.
The Raspberry PI, analog "hat", display and sensor wil probably cost more than $500 and then there is "some assembly required"
I doubt you can do what you want to do for $500 and if you can, it will cost thousands of $ of your time to make it so.
Another issue that people haven't brought up is that the analog sensor and analog input need to be fast to catch those 6000 psi spikes.
Something that might help is to attach a small accumulator that is pre-charged to above system pressure. If a pressure spike occurs then
the accumulator will absorb it but otherwise not affect anything while the pressure is below the pre-charge pressure.

 

[3DDave]

I am tempted to suggest a relief valve and a bucket. Set the relief some amount above the typical pressure and then measure how much fluid ends up in the bucket. The bucket could be more of a jar or other container that won't get sloppy. Put a sintered filter to keep the jar from getting pressurized, for example.

If the goal is to prove the pumps are failing when there aren't any pressure spikes, the lack of leakage into the bucket would be as suitable as sending a few gigabytes of data. Either the pump maker will back their product or they won't. If the pumps are failing now, what are they supposed to do to make them not fail in the future?

Perhaps it is time to get a different pump supplier.

 

[PNachtwey]

If any of you can suggest a device that meets the criteria above, I would appreciate it.
Thanks,

Since no one has come up with a solution.
Our cheapest analog motion controller can to that easily but it will blow your budget. It can capture data at rates as short as 250 micro seconds and can be configured to trigger on events so when the pressure spike occurs, you won't miss it an the data can be logged to you can see what led up to the event and what happened other. This is similar to what the old logic analyzers did. It can interface with a laptop use Ethernet or serial but Ethernet is better for monitoring data in real time. The serial version would do but the laptop would need to use a serial to USB adapter.

 

[HydroBuff]

a relief valve and a bucket

Cool idea! I actually thought of the same, but decided that it really didn't look very professional.

Our cheapest analog motion controller can to that easily but it will blow your budget. It can capture data at rates as short as 250 micro seconds and can be configured to trigger on events so when the pressure spike occurs, you won't miss it an the data can be logged to you can see what led up to the event and what happened other. This is similar to what the old logic analyzers did. It can interface with a laptop use Ethernet or serial but Ethernet is better for monitoring data in real time. The serial version would do but the laptop would need to use a serial to USB adapter.

Thank you for your suggestion. I actually already have access to IQAN plc's and will easily be able to accomplish the same thing with them, only I didn't want to wire up and go through the work of programming etc. if there would already be something available off the shelf.
I was thinking more down the line of a ready made Parker/ Webtec etc. If anyone sees this and has any suggestions, let me know.

 

[Stick.]

I was thinking more down the line of a ready made Parker/ Webtec etc. If anyone sees this and has any suggestions, let me know.

I think anything off the shelf will blow your budget (it's pretty tiny, particularly given the apparent magnitude of failures you're seeing), but you could always contact the sales departments at Parker, Webtec, etc and find out.

I don't know how much granularity you need to your data, how often you can monitor it, etc, but you could try out a mechanical gauge with a max pressure needle and just check it daily/weekly. Not a great solution, but it's at least under budget.

 

[HydroBuff]

I think anything off the shelf will blow your budget (it's pretty tiny, particularly given the apparent magnitude of failures you're seeing), but you could always contact the sales departments at Parker, Webtec, etc and find out.

I should clarify on my budget. I threw out the number $500 because it would have been cool to be under that.
We are, however; a small company and I was not handed that budget from anyone above, if I say it is neccesary, I can get it. I am more than willing to pay more if neccesary.

 

[3DDave]

Let's say you have the data. Then what?

You have a pump failure problem, not a data capture problem.

 

[HydroBuff]

Let's say you have the data. Then what?

You have a pump failure problem, not a data capture problem.

I would say we all know we can't have too much data when analyzing a failure.
I have not been able to pinpoint the failure cause of these pumps yet, although I expect that it is most likely a pressure spike at random times. I was not able to prove that yet, thogh.

 

[3DDave]

A pressure spike from what? The pump is the source of the flow; restriction is the source of the pressure.

What part of the pump is failing?

 

[HydroBuff]

A pressure spike from what?

To be honest, I can't prove that is what is happening which is why I am investigating the possibility of a data logger.
We are using a Parker tripple gear pump, and the section that is running the main boom of a truck loader is seeing various failures.
Some pumps crrack the housing, others scar the gear housing, and others have a complete bearing failure. While it is remotely possible that it is contamination, I feel I have investigated enough into that to start looking elsewhere for the cause of failure. We are using F130 valves, so that should not be the problem, but I need to prove it anyway.

 

[3DDave]

Where are the relief valves?

Is there a system schematic?

 

[HydroBuff]

The relief valves are installed on the F130 valve.
The system schematic is proprietary, let me see if I can cut some snippets that show the neccesary functions.

 

[3DDave]

Perhaps the relief lines are restricted. I've seen that a couple of times, where performance was substantially restricted because return lines were grossly undersized. Like, putting 90% of the supply pressure onto the relief line. In one case this blew out the seals on motors from restricted case drains. In another case, instead of looking at the system, they added a secondary relief line, with a solenoid valve, and powered the valve from the power supply to the anti-lock braking system. Same result would have been had with a move from a .38 inch return line to a .63 inch return line. That simple.

 

[HydroBuff]

Perhaps the relief lines are restricted.

That's what I'm highly suspicious of. For instance, when the boom is coming down with a heavy load and multitasking.

 

[3DDave]

The pump should not be exposed to spikes when coming down. There should be a counterbalance valve that the pump is driving open to produce make-up fluid to the unloaded side of the cylinder(s.) The pressure should be dropped from outflow from the cylinder by the valves, not fed back to the pump.

The test is simple. Block the outlets from the valve, except the relief port, and floor the gas pedal and see if the pump breaks. If the pump breaks you have insufficient capacity in the relief circuit. Don't do this long as all the energy going into the fluid will be coming out as heat; not ideal. But there is no doubt a temperature sensor in the reservoir to disengage the PTO if the fluid gets too hot, right?