Calculating Heat Input - Variable Polarity TIG?

Date 03-19-2010 22:34

Heat input [J/in. (J/mm)] =
Voltage * Amperage * 60
/Travel Speed [in./min (mm/min)]
this formula from asme IX
but i don't know if it will help you .
Regards
Mohamed Fatehy

By ssbn727 (*****)

Date 03-19-2010 23:55

Hi Zeek!

When you're using variable polarity, things aren't "standard" anymore! ;) The same goes for when you add pulsing into the formula as well because of pulse peak and dwell times which means you must find the average heat input which can vary very little to a significant amount depending on your power source polarity and/or pulse settings. So, a slight "tweek" in the "standard " formula is necessary in order to accommodate these other variables, and to ensure more accuracy IMHO! ;)

Respectfully,
Henry

By Metarinka (****)

Date 03-21-2010 17:54

now the formula becomes much long and interesting. You need to find how much heat input you putting in at DCEP and DCEN your current and voltage are identical, so the difference is from electron direction. Once you find that heat inputs of your two polarities you. You take the weighted average. This will give you the average heat input per second.

you'll have to do that calculation twice, once for your peak pulse current and one for your background pulse current. then once you have the weighted average of that you'll have the average heat input per second.

very confusing to say it verbally.

IF the AC is a sign wave technically you have to do it with calculus, but you can use an approximation

By MBSims

Date 03-21-2010 18:27

I guess you all lost me on the need for calculus or complicated math for the AC portion. As far as I know, the traditional Heat Input = Volts x Amps x 60 / Travel Speed equation is just as valid for AC as it is for DC. If I take 85% x 230 for the DCEN and 15% x 230 for the DCEP, it still adds up to 230 amps.

The missing info for the pulse portion is pulse width for primary and background pulse, and the background pulse amps.

You could use the Power Ratio heat input approach which is: Power Ratio = (Amps x Volts)/[(wire feed speed/travel speed) x cross sectional area of wire]

By Metarinka (****)

Date 03-22-2010 14:59

As we all know heat input is different depending on the polarity of the torch/work. The overall energy is the same for AC/DC but more or less is going into the work or torch. Running AC at 230 Amps is not equivelant to running DC at 230 Amps. The classic formula (V x A )/ Travel speed; is perfectly acceptable to gauge heat input until you talk about switching polarities. Really it depends what you're trying to measure, heat input into work, heat input in the electrode, or current capacity of the wire.

If you have an off balanced AC wave your heat input will be some value in between the two polarities. now if it's a true sine wave, since at any moment the arc is at a different current, technically you need to use calculus to figure out the averaged arc current as heat input is offbalanced between the 2 poles. But that's a little bit of overkill in the accuracy department.

what I was trying to hint at is that the classic formula doesn't account for the fact that DCEP and DCEN provide different amounts of heat input into the work piece for the same wattage. If you use that formula for an off balanced AC wave, then your heat input into the work will appear to be higher than it actually is.

By js55

Date 03-22-2010 17:03

I think the discussion of theory is a good one, but from a practical point of view the difference in heat input between AC and DC is a 'who cares'.
If you switch polarities in an impact regime you have to requal anyway, its a Supplemental.
The proof is in the testing puddin.
If you get 50kj with AC and 50 kj with DC using the standard formula it doesn't matter if the energy imposed on the weld is the same or not.
You gotta test it either way. You can't translate from one to the other without requal.
If you test and it tests good, run it at a 50kj limitation.
If you have a concern about phase balance then impose AC phase balance limitaitons on the WPS.
The code requirements are minimum standards not research level metallurgical verification mechanisms.

By Metarinka (****)

Date 03-23-2010 15:41

very true, it really is more of an academic distinction. I run my formulas that way more so I can compare heat input during polarity changes within a process. Helps me estimate variables without having to guess as much.

By Lawrence (*****)

Date 03-21-2010 18:40

Zeek

Are you running a pulse frequency of 80pps.... If yes give dwell details.

Or are you running an AC frequency of 80 Hz.

If you answer this question I think the other guys will have you on the right track snappy quick.

By Zeek

Date 03-22-2010 19:01

Lawrence -

Here's what I'm running
80 pulses per second
85% peak time
85% background amps

The frequency is set at 400Hz in order to get a much narrower arc. Does that help at all?

Thanks.

By MBSims

Date 03-23-2010 00:10

Let's see if I remember this correctly:

Heat Input = Volts x Amps x 60 / Travel Speed

For a weighted average it would be:

Heat Input = Peak Heat Input + Background Heat Input
   = [Volts x (Peak Amps x Pulse Width) x 60 / Travel Speed] + [Volts x (Background Amps x Pulse Width) x 60 / Travel Speed]
   = [10.6 x (230 x 0.85) x 60 / 13.5] + [10.6 x ((230 x 0.85) x 0.15) x 60 / 13.5
   = 9210 + 1382
   = 10,592 J/in.

By MMyers

Date 03-23-2010 20:16

Sounds like you're running AC with Pulsing. IMO, you may as well go home because you're in for a nasty equation because me thinks the waveform will not be stable, but instead an interlace of 400Hz and 80Hz, so your actual waveform changes as a function of time. You should be able to hear this as the wave length will repeat every 5 seconds. Verify this with an oscilloscope if you have one, I'm shooting from the hip.

By MBSims

Date 03-26-2010 13:25

In the end it's only a way of evaluating "changes" in heat input. To get a number that is even close to the real heat input, the thermal efficiency of the arc for the process used would have to be considered. Thermal efficiency of the arc is not a part of the traditional heat input equation, since it is only used to compare changes for a given process and polarity, which are both supplementary essential variables (as Jeff mentioned). Fluke makes a meter that can be used to measure "power" where waveforms vary. Such a measurement could be inserted into the "volts x amps" part of the equation to get a more accurate number for heat input.

http://us.fluke.com/fluke/usen/Power-Quality-Tools/Single-Phase/Fluke-43B.htm?PID=56080

By ssbn727 (*****)

Date 03-26-2010 15:45

Like I said previously, a slight "tweek" or addition in the standard heat input formula is all that's necessary to figure out the heat input average. :)

Respectfully,
Henry

By MBSims

Date 03-30-2010 01:49

Got scrap? Go for it!

From http://www.microwaves101.com/encyclopedia/slang.cfm:

Kentucky Windage

A straightforward and effective way to recenter a design that is off in frequency. For example, you design a filter to work at 10 GHz center frequency. In test, it's centered at 10.5 GHz. Instead of analyzing the crap out of what happened, Kentucky windage means you redesign the filter for a center frequency of 9.5 GHz and expect it to work at 10 GHz on the second try. This term dates back to Sergeant York, a sharpshooter from Tennessee in W.W.I., who was an inspiration to gun nuts all over the world. Often employed in antenna design.

By 803056

Date 03-30-2010 04:42

Hook the machine to a multichannel oscilloscope to see what the actual waveforms looks like and calculate the "area under the curve" for voltage and amperage.

The heat input equations as we typically use them are an approximation that overstates the actual heat input. Each welding process has an efficiency factor (that is usually omitted) that is used when "correcting" the heat input to provide a better estimation of the heat input (Q).

Lincoln has a program that allows you to hook up to their PowerWave power supplies that turns your laptop into an multichannel oscilloscope to see the actual wave form and to determine the heat input of the machine operating under different modes. Unfortunately, the program can’t be used with other machines because there is no connection provided by the manufacturers.

I always wanted to try one of Fluke’s Scopeometers (like the 199C), but could never justify the cost. Then there’s the though that the high frequency output for GTAW could possibly fry the test instrument the first time it is connected to the welding machine. Oops, there goes another $3000.00!

Best regards – Al

By ssbn727 (*****)

Date 03-30-2010 05:40

Hi Al!

I used that program a few time already and it sure does work like you describe... Too bad it's proprietary to Lincoln power wave products exclusively! ;) I would use a scope for other brand power sources like you mentioned also... Heat input equations usually turn out to be approximations because of the fact that not every variable is considered as part of the equation, so I agree that at best, one can come up with an estimate that is only as good as the appropriate variables being included or not! ;)

The newer Fluke meters have protective circuits in them designed to blow a fuse if such a thing happened so, you'll be protected. ;)

Respectfully,
Henry