|Ian Parkin||21/08/2015 11:50:49|
1016 forum posts
Does anyone know if such an instrument as a air ammeter is available?
Presumably a anemometer
In my day job I'm called out to large machines that have a lot of pneumatics on them.The poor compressor is running continuously and the operator reports "we have air leaks"
So have to start removing panels and looking listening for air leaks can take ages particularly when theres multiple small leaks.
These machines are large multi unit printing machines with a common air line running down the length of the machine each unit has a manifold with valves for each unit
If I could put an "ammeter " into the airline coming from the compressor I could clamp the supply line at various places and narrow down where the major leaks are before removing panels to the various units
Something like the flow meter (with the ball rising in a tube ) fitted to welding bottles but safe up to 150psi
My local air components supplier cant help but suggests a ultrasonic listener device at vast expense which needs panels removing and relative quiteness...not easy in a large commercial printers whilst everyone else is running.
anyone any ideas?
1808 forum posts
I think you are looking for a 'Manometer'
|3546 forum posts|
Hire and try leak flow meter
|Ian Parkin||21/08/2015 12:00:56|
1016 forum posts
doesn@t a manometer measure pressure
I need to measure flow
like volts equals pressure amps equals flow
1808 forum posts
Yup. Sorry, my mistake.
|Neil Wyatt||21/08/2015 12:06:15|
18990 forum posts
You need a MAF sensor (mass air flow).
Every modern car has one on its inlet manifold and I am sure you could hack one. They usually use two heated wires, one shielded from the air flow the other not and compare their resistances
You could even use the more primitive style as fitted to the old Astra GTE about 20 years ago - just a loosely sprung flap attached to a potentiometer.
Just noticed your pressure requirements!
This goes up to 100 psi.
Google gives lots of hits for compressed air flow meter
Edited By Neil Wyatt on 21/08/2015 12:08:41
|Ian Parkin||21/08/2015 12:22:16|
1016 forum posts
That seems to be the sticking point the pressure requirement
they all seem to go up to 6.9 bar or 100 psi
sometimes the machines run at 120-150 psi
I wonder if a flow meter states safe to 100psi what will happen if its 50% more
It could explode showing vicinity with shards of perspex. I could enclose it in a clear polycarbonate tube vented
I have ordered a flow meter off ebay I'll try it at 150 psi for a while to test it
|Martin Kyte||21/08/2015 12:31:06|
2721 forum posts
If your flow rates are reasonable you could install a restriction in the line and measure the differential pressure. Just the same as measuring the voltage accross a sense resistor to measure current in an electrical circuit. Should work for a reasonably steady flow. If there is a lot of 'pulsed load' it's going to be a little difficult to see what is down to you shutting valves and whats not.
|David Jupp||21/08/2015 12:42:49|
|822 forum posts|
Ultrasonic leak detectors are surprisingly effective in noisy environments, the 'noise' is typically in a very different frequency range from that which leaks give rise to.
Rather more expensive - you can now buy clamp on ultrasonic flow meters that can measure gas flow rates.
A simple device that seems to offer what you ask for would be a Rotameter - a 'float' rises up a graduated tube , more flow, the 'float' lifts further up the tube. A quick search on Rotameter should show some options.
|Ian P||21/08/2015 13:14:08|
2578 forum posts
'Float in tube' Rotameter type instruments generally have a limited range of measurement, there is a big difference between measuring flow in mililitres per hour and cubic metres a minute.
Leak testing the sort of system you describe should be done in a logical and sequential manner. I doubt whether you could do it whilst the machine is running and various valve etc are operating.
With the machine in a static state I presume it should not be consuming any air? If you then insert a flowmeter in the compressor output line and there was only a small leak it might not show any reading unless it was a very low range flowmeter. You only really know what range of measurement you need 'after' you have measured the flow!
Any flowmeter that will handle the flow of the equipment is unlikely to even move off its zero position with just the leak rate. Unless the machine is very cooperative even inserting a flowmeter into the line can be very trying or inconvenient because of the various pipe and fitting sizes.
I would start by isolating sections at a time and use the soapy liquid type indicator.
|Ian Parkin||21/08/2015 13:37:17|
1016 forum posts
Yes you are right the machine should not be consuming any air whilst at rest or indeed whilst running
only when certain operations are called for but for most of its running time its drawing no air.
But the poor compressor is running 100% dealing with leaks
All I need is a device to go into the airline and tell me if theres any flow, I'm not really interested in the flow rate just if theres any flow
so I bought a couple of these to experiment with
Most air supplies are 10mm or 12mm nylon tube so it wont take long to adapt
if you can imagine a machine 40 ft long with 6 units and each unit taking perhaps 30 - 120 mins to take its panels off any time saving is good
If the device i've bought stands up to the pressure then the ball will be at the top of the column so by crimping /isolating pipes downstream of the meter will quickly show where the major leaks are
I've been fiddling with printing machines now for 35 years so I do work in a logical fashion but with machines getting more automated with electronics and air actuators I could do with some help with diagnosis as they get older and more leaky
Lots of times the source of the leak is rodents they seem to like the taste of nylon tube
|Ian P||21/08/2015 14:19:58|
2578 forum posts
As I dont know your system i have no idea how big the compressor is but if its running continuously there must be a fair leak, Can you isolate each cabinet in turn?
'Crimping' nylon tubing enough to stop the flow might destroy is integrity especially if its aged, its better to disconnect and cap off.
The flowmeter you show will give no indication below 0.4ft3/hr which to me sounds like a big leak (but its all relative anyway)
|Clive Hartland||21/08/2015 14:25:15|
2810 forum posts
If you are looking for air leaks then something like a foam spray will show if there is a leak. One thing to check is if there is any water in the system in pipes that are sagging down anywhere, ( I have had that happen) It is no good having moisture traps if every time you bleed one you get a half pint out of it. The compressor running 100% of time will give quite warm air which is ideal for making condensate.
Pinching a tube to measure flow? You are seeking how much flow, so you need a measurement at full flow. You might as well buy an Anemometer and take a tube off and direct it at the vanes. Check spec of air consumption for each take off point, you might find one part is at fault.
|Ian Parkin||21/08/2015 15:21:03|
1016 forum posts
0.4ft3/hr would be nothing i'm talking 4 ft3/min
these machines are massive 30-40 tons running maybe 10 compressors (lp)
and one HP compressor maybe with a 50-100 litre tank
I know it seems strange to say that you cant hear the leaks but you cant.... taking panels off to look with a spray just takes too long
heres the layout of a typical press
the air supply pipes genrally run at ground level down each side of the machine
So if I place my meter at A I can then isolate at E,D,C,B on each side and look at the meter to see which unit is leaking most before stripping down further
then further on each set of pipes coming from the manifold maybe 30 pairs
there may be 30 actuators in each unit
Some presses you can spend 3 hours removing just one side panel and then similalr putting back together
For anyone who knows these are KBA's or man rolands
heidelburgs are a bit simpler
|John McNamara||21/08/2015 16:09:20|
1336 forum posts
Provided there are no electronics nearby an easy way to test for leaks is a plastic trigger spray bottle filled with water mixed with plain soap. Standard equipment for Gas fitters. You don't need much just a light spray on any joints or suspect areas.
Even a tiny leak is easily shown by the bubbles.
|Ian Parkin||21/08/2015 17:13:08|
1016 forum posts
John what people are not grasping is that to take ALL the panels off takes perhaps 2 days to find a leak that takes 2 mins to cure when the panels are off the leaks are easy
I need to speed that up by diagnosing where it may be before dismantling
This is a multi million pond press new and with 3 very highly paid blokes not working so time is of the essence
|frank brown||21/08/2015 19:17:29|
|436 forum posts|
The only way I can think of is to install good self sealing air connectors T'ed between the sections, then you could use some form of differential pressure indicator on a test lead. Either a sensitive pressure meter whose works can withstand the static pressure and a over pressure valve across it. This is because when a solenoid operates the pressure at far end might drop by 50 PSI and you don't want to blow up a 0-10 PSI meter. A turbine in a pressure rated case would do, if there was air going through it, it would rotate slowly. This could be done in a metal case with the blade rotation indicated by the blades cutting a light beam to a photocell, so it would either "tick" or "scream" depending on the air flow.
|Keith Long||21/08/2015 19:37:14|
|876 forum posts|
Ian - you might find something in the variable area flowmeter range made by Platons (http://www.ctplaton.com/en/home.php ). The working pressure that I saw was up to 10 barg so in line with your supply pressure. Be worth enquiring by phone or e-mail. I use to use these devices, and similar, a lot 30 years ago on both liquids and gases, usually very easy to set up and operate - just needed to get the range right, we had a kit of interchangeable parts that allowed you to swap tubes and floats to cover a large range.
1342 forum posts
As I understand it, you have:
1. A compressor that should be supplying 4ft^3/min of air at 150 lbf/in^2.
2. One or more printing presses that have air leaks
3. A management that does not want to shut down the printing presses - I guess that this will be the real problem.
First, do the easy thing. Check for air leaks from the compressor to the presses, that is all the flanges etc, using soap and water or a proprietary spray (both have their advantages). Rectify any leaks.
Second, if you are still unhappy you will have to shut down the process and install some form of air flow measuring device at the delivery of the pump (A on your sketch). I do not know what devices you have found on the internet but at work, where we had to measure similar air flows, we used British Standard BS1042 designs (usually D to D/2 orifice plates with suitable pressure transducers). However you may not have the room to install the correct pipework. If you do look at BS1042 do not be put off by the size of the document. Now start up the whole system and shut one press down in turn measuring the air flow each time. You can compare the flow for each press with its specification and come to a conclusion which press is leaking (or leaking badly) and take appropriate action. Of course it is possible that the compressor is unable to supply the required amount of air.
As said earlier the real problem is the financial benefit of keeping the presses running against the cost of producing excess high pressure air.
|Andrew Entwistle||21/08/2015 19:49:08|
105 forum posts
I am not very knowledgeable in this area, but we just installed one of these datalogging flowmeters at work to meter the compressed air used by a company we share the building with.
It works at up to 16bar. It must be possible to get a much cheaper non logging version (this one was about £1k).
Edited By Andrew Entwistle on 21/08/2015 19:49:36
Please login to post a reply.
Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!
You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.
Click THIS LINK for full contact details.
For subscription issues please see THIS LINK.