Ammeters

[Hitch]

With this CE marking stuff coming in, it seems we need calibrated ammeters on all the machines....

We have 4 with already, so no big deal.

With the others, i can't see that fitting a panel mount ammeter onto the machine should be all that difficult, either on the power source, or on the feeder. These would all be transformer based machines.

Where would the ideal locations to connect be, one wire on the return, one on the torch connector? Is it that simple?

Analogue or Digital... i imagine an analogue one is easier and cheaper to install?

 

[AndyStobbs]

I've just been looking at them on ebay. Digital DC seem very inexpensive. You could connect on the earth lead or the power lead as current is conserved.

The ones I've been looking at seem to be either a shunt or hall effect for current sensing with external power source for the circuitry.

 

[rtbcomp]

........
Where would the ideal locations to connect be, one wire on the return, one on the torch connector? Is it that simple?

..........

That's where you'd connect a voltmeter.

An ammeter must be in series with the load, i.e. you'd disconnect the return and connect that wire to the ammeter and the other connection on ammeter goes to where the return was connected to.

You could use the torch feed if that's more convenient.

 

[wheresmejumper]

That's where you'd connect a voltmeter.

An ammeter must be in series with the load, i.e. you'd disconnect the return and connect that wire to the ammeter and the other connection on ammeter goes to where the return was connected to.

You could use the torch feed if that's more convenient.

is there a 'hands free' type exist like a clamp meter?dont know

 

[Jim Davey]

Analogue can be easier to read if there is a little instability in the arc as the needle will oscillate around a small area of the scale, with digital units a rapidly fluctuating number can be hard to read. Plus points of digital is they're relatively unscathed by dust and grit that can affect moving coil analogue gauges, and being illuminated they can be easily read in low light.

Meters themselves are relatively inexpensive, the cost is usually tied up in the shunt, which is often a sizeable lump of copper or brass. Bear in mind the shunt will need fettling to get it to provide the correct amount of millivolts to the meter, some are 50 some are 60 so buy with care. Also bear in mind that the shunt has to take full welding current of the machine so select one with a duty cycle equal or better than the machine you want to put it in.

For more info on calibration of arc welding equipment look up BS/EN5750:2008.

 

[rtbcomp]

is there a 'hands free' type exist like a clamp meter?dont know

AC ones would be cheap enough, DC not so.

 

[Robotstar5]

Once fitted, you then have the annual cost of getting the calibration checked

 

[Justme]

A shunt is not going to last long in a welder unless it is massively over rated.

You would be better with a clamp on / induced type.

Not as accurate but will last longer.

A digital one will need some sort of damping effect so that it can be read.

The one built into my welder adjusts whilst welding & when you stop it holds that reading for you so you can see it.

 

[Jim Davey]

A shunt is not going to last long in a welder unless it is massively over rated.

Not at all, jut buy a shunt that's adequately rated for the load intended, I've got boxes of them here going right up to 600 amps.

 

[the snooper]

are these ammeters intended to set the welding current correctly while setting up or just monitor the output, its going to be hard to do while using the machine

 

[Wedg1e]

Not at all, jut buy a shunt that's adequately rated for the load intended, I've got boxes of them here going right up to 600 amps.

Pah, 600A, you call that a shunt?

I've just been calibrating some Mag Particle Inspection machines in Dublin (posting this from the ferry home), the three shunts I have are 1000, 2000 and 6000A! Would make some pikey's day if I left them lying around somewhere, they are big blocks of copper...

 

[Jim Davey]

Nice, the biggest set I see down my way is a 1200 amp Lincoln that gets used for gouging. It's all stainless, ally and tin down south, we prefer to drink posh coffee and work on our tans than mess around with heavy metal

Seriously though, there are a few big sub arc systems in my patch but they are looked after by another firm, I've never even come across an application for it, there just isn't the heavy steelwork going on on the central south coast that there was a few years back.

 

[Wedg1e]

Nice, the biggest set I see down my way is a 1200 amp Lincoln that gets used for gouging. It's all stainless, ally and tin down south, we prefer to drink posh coffee and work on our tans than mess around with heavy metal

Seriously though, there are a few big sub arc systems in my patch but they are looked after by another firm, I've never even come across an application for it, there just isn't the heavy steelwork going on on the slcentral south coast that there was a few years back.

A Mag Particle machine is almost identical circuitry-wise to a welding set: essentially a big (BIG) transformer, rectifier if required (some sets are AC-out only, some are switchable AC/DC and some are DC-only - then you can have half- and full-wave rectified!).

I used to do some welding set calibration but it wasn't what you'd call scientific: essentially I hung a couple of clamp ammeters and a voltmeter on the output, then got the operator to fire up the torch/ strike a rod at various settings while I took the readings. It was quite amusing watching the welders competing over who could get the steadiest run - the really low settings with stick were the trickiest as I recall.

The client in question just wanted a 'certificate' to wave at potential customers; they weren't bothered about the last degree of accuracy that would be demanded nowadays with NAMAS, UKAS, ISO etc.
As I always say: it doesn't matter if the job's done right as long as the paperwork says the job was done right...

MPI and welding machines are calibrated the same way: hang a test load of a known quantity on it and press the start button. Here the load is a shunt resistor of a fraction of an Ohm - but with 6000A passing through it, it's some hellish wattage! An extended period of testing can mean you can't touch the biggest shunt for an hour until it cools - and I've melted a good few lengths of welding cable on MPI machines over the years

 

[Jim Davey]

I'd get hung if I was caught using open arc to calibrate a set. It does **** me off that people are getting fleeced for calibration though, some guy wanders round with a multimeter and a clipboard and charges £15 a set and can do 100 machines in a day and all it is a complete fabrication as the machine was never under load. Following the British standard I'm pushing it to do 8 a day, that's if they all go smoothly and none need adjusting up.
Problem is that the customer perceives that that's the cost of a calibration and falls off of his chair when you tell him a proper one is more like three times that price!

 

[matt1978]

I'd get hung if I was caught using open arc to calibrate a set. It does **** me off that people are getting fleeced for calibration though, some guy wanders round with a multimeter and a clipboard and charges £15 a set and can do 100 machines in a day and all it is a complete fabrication as the machine was never under load. Following the British standard I'm pushing it to do 8 a day, that's if they all go smoothly and none need adjusting up.
Problem is that the customer perceives that that's the cost of a calibration and falls off of his chair when you tell him a proper one is more like three times that price!

Couldn't agree more. Too many cowboys wacking calibration Certs out when they have no way done it properly.

Be interesting to see what happens with this EN1090 because through the different specs you can trace it through that the machines are supposed to be Calibrated in accordance with ISO EN 50504, and there is no way you can calibrate a machine to that spec in 10 mins, nor can it be done with a normal run of the mill static load bank.

 

[Wedg1e]

Couldn't agree more. Too many cowboys wacking calibration Certs out when they have no way done it properly.

Be interesting to see what happens with this EN1090 because through the different specs you can trace it through that the machines are supposed to be Calibrated in accordance with ISO EN 50504, and there is no way you can calibrate a machine to that spec in 10 mins, nor can it be done with a normal run of the mill static load bank.

Part of our business is instrumentation calibration (not welding kit): most 'calibration' is like the MOT on a car, only valid for about 15 seconds and only when there's a Z in the day.
We spend a fortune having our reference standards and lab instruments 'calibrated' by other parties, all anyone has to show is a paper-trail a mile long with no guarantee that anything's more accurate than it was before.

Probably the single most pointless 'calibration' we do is the focal spot dimension check of an x-ray set. It was Rolls-Royce who started it and now almost anyone who is a component supplier to them has to have this test carried out annually - or in some cases 6-monthly. We've done tests on 40-year-old x-ray heads that showed no measurable deterioration so how anyone expects to see a change in 6 months is a complete mystery... but the customer is always right so we turn up and do the measurement

Our boss is keen for us to attain UKAS accreditation in three particular fields, one of which in particular is so open to end-user (mis)interpretation that it will be completely pointless. The cost to the end users will have to rise to cover the additional labour necessary: it'll be interesting to see how many clients forego the UKAS cert. and go elsewhere to save money.

 

[Jim Davey]

Did you say you can't test to EN50504 with a load bank?

 

[matt1978]

Not that you can't use a load bank but a " static" load bank. AFAIK you need a variable load bank that can change its resistance properties according to which process your testing.

As you know there are different parabolic curves as to whether its Mig, MMA or Tig and if a machine is capable of more than one process such as most inverters,are at least Mig and MMA your supposed to have the means of checking the load on each process to adhere to the EN 50504 spec properly.

I am no way an service engineer so correct me if I am wrong but I did see a presentation on it in relation to EN 1090 and I am sure that was pretty much the gist of it. I'll see if I can dig it out and read through it again.

 

[Jim Davey]

Ah I see, I must admit, I've not been made aware of the full extent of the requirements for EN 1090, I do use a variable load bank though so I'm ok on that front.

 

[Jim Davey]

For more info on calibration of arc welding equipment look up BS/EN5750:2008.

I put the wrong number there

The correct number these days is BS EN 50504:2008