TIG Pulser for Miller 180SD

Page 1 of 2  
I've been putzing with a design for an add-on pulser for my Miller 180SD TIG welder at work. After 3 years, on-and-off (mostly off), I've finally got a working unit. I posted it under "TIG Pulser" on the Dropbox:
http://www.metalworking.com/dropbox /
I finished it today, so I can't give you any long term use information. But, it seems to work as designed.
Many thanks to Gary Campbell, Ray Sommer,....and especially Ernie Leimkuhler for his willingness to share all his welding experience with us.
Ken





Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Looks pretty cool.


Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Thanks Ernie!



Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Interesting, Ken! What benefit does it provide you while welding? I'm in the Twin Cities, might I have a look at it and try it?



Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I've been wondering this also. Two per second, I think I understand: Sets a fixed interval for the heat, fill, move cycle. 20 per second? What is the benefit? That seems certainly faster than I can see.


Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Don Foreman wrote:

Pulsing helps ensure penetration on thin materials without burning thru. To try it just pump your foot pedal. Not as controled as a pulser, but you will see how it helps.
Dan




Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

180SD?
to
for

A pulser is a device used to interrupt a welding arc. Mostly used on TIG, but can also be used for MIG. The idea is to cycle the arc from high amperage to low amperage. The metal melts at the high amperage and solidifies at the low amperage.
There are 3 settings for a pulser. 1. Background amperage This sets the low amperage and is usually set as a percentage of the main amperage. So if you main amperage is 100 amps, and the Background Amperage is set for 50%, then you Background Amperage is 50 amps.
2. Percentage of On time. This sets how much of each pulse cycle is spent at the high amperage and how much a the low level, once again by percentage. A setting of 50% evenly splits the cycle between low and high amperages. Less than 50% on On time gives a Spikier Pulse. Greater than 50% gives a Softer Pulse.
3. Cycles per second. This is the really confusing one, because it has the most profound effects. Older TIG pulsers only allow up to a maximum of 10, or less, Cycles Per Second (Hertz - Hz), but the newer Inverter TIGs allow much higher frequencies. My Maxstar 200DX goes up to 200hz.
The lower pulse frequencies are for traditional pulser use, where you run between 1 and 2 hz on the pulse. adding filler metal on every pulse or every other pulse as you progress across the weld.
The higher frequencies have a much different effect, in that they tend to make the metal super-liquid, allowing it to flow and wet out much smoother. These higher settings work very well with autogenous welds where no filler metal is added at all. These are also called Fusion or Flow welds. I also found these higher frequencies worked well for vertical welds in heavy Silicon Bronze.
Now the reason for all this pulsing is rather simple. Lets start with a piece of 16 ga steel. At 0.062" thick the proper amperage would be 1 amp per 0.001" of thickness or 62 amps, for a flat-butt weld, full penetration, single pass. You could easily TIG weld a seam in 16 ga Steel using a continuous 62 amps. However a spike of high amperage will melt the metal much faster than a low amperage, so you could weld faster at 80 amps than 62 amps, but at that heat you risk overheating the metal and causing undue warpage or burning of the steel, so you mix an interval of 80 amps with an interval at 40 amps. The low amperage interval allows the puddle to solidify back into steel, without allowing the arc to break. By pulsing the weld across the bead you get a very orderly row of rings, that give that distinct stack-of-dimes look to the weld bead. You adjust the exact pulse frequency to suit your style and speed of welding as well as the thickness of the metal. The other benefit is that over all, you have put less heat into the metal, so you have less distortion.
A smaller, consistent weld bead is often stronger than a larger inconsistent weld bead.
The more consistent a weld is the more the stresses apply along it's entire length. Any peak or valley in a weld becomes a stress point for failure to occur.
Whether you choose to add filler metal on every pulse, every other pulse, every third pulse or not at all is dependent upon your own welding style and the circumstances of the weld.
Where pulsers become bewildering is when you start messing with the percentage settings for the Background Amperage and percentage of On Time. The number of possible combinations is huge, and there is little or no guidance given in the welding world as to application.
Pro-Fusion has an excellent series of webpages with online calculators that allow you to punch in the overall required amperage you want and how fast you want to weld and it will generate pulser settings for you.
http://www.pro-fusiononline.com/welding/pulserate.asp http://www.pro-fusiononline.com/welding/pulseparams.asp
It still gives no guidance as to when you want a Spike pulse, or a Soft pulse, but it gives you something to play with.
50%, 50% and 1.4hz, is a setting that works for most sheet metal.
All I can say is that until somebody puts out some really authoritative research showing what pulser settings are best for what combination of weld and materiel, we will all be out there experimenting on our own.
I have searched for such a book for 5 years, and given up. Hopefully some day soon the welding engineers will decide to enlighten us lowly welders.
The guideline I can give you is that the thinner the materiel the more it benefits from a spikier pulse. I have yet to find a benefit to a really soft pulse.
An example of an extreme spike pulse is that you can weld a popcan at 120 amps, as long as you set your pulser to 2% On Time and 2% background amps.

























Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

That got my attention! This project just went on my "to do" list. I probably have all the parts on hand, except maybe for the pots. Ax Man has pots.
I'd be interested in one of your PCB's if you still have any left. Most folks would do this with a PIC processor, but your linear design is clean, straightforward and very un-fussy. Probably also pretty EMI-proof, and no higher chipcount than a UP would need.
What is that block in the lower righthand corner of the schematic, pls?





Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Don,
I still have the PCB's. The Log-taper pot was the only one I had to buy. I had been through Axman's stock with no luck. ;) I'll look through my old PO's and get you the vendor and part #.
I had thought about a PIC halfway through the developement. But the HF arc starter was blowing out the LM324 on one of my perf-board prototypes, and I didn't want to get into that noise screwing up program steps. The addition of a ground plane on a PCB and the ferrites and bypass caps on the wires entering the box solved that. So, that may be the way to go now. Though the analog verson did everything I wanted.
The block is a DC-DC converter, to get a bipolar supply (+/-12vdc) for the op-amp, from the +20volts in the welder. It needed that to drive the P-channel MOSFET into full conduction. It was a 12vdc input DC-DC converter that I already had...that's why I had to put U1 in there...to get the 20v down to 12v. I can't find one with a 12v input that has the same footprint and pinout to directly drop in. That's one of the mod's on the "Mark II" if I ever did another one. ;)
Ken






Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Amazingly enough, the programmable (in BASIC) controller in my own modded welder, called Cubloc, is not burning out from HF.
i



Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

The early prototype didn't have any provision for an HF arc'y environment. T'was a learning experience.
I was looking at a PicAxe-08 (also in BASIC) at the time, but not enough ADC inputs. Now the -08M has 3 ADC inputs..for the pots. Unless this unit goes up in smoke, or some other "must-have" feature comes along I'll stop here and just enjoy using it.
...well there is a Slope function that might be useful! ;)
Ken





Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

The modern stuff that is sold these days has some interesting features. One is more analog inputs. Another, is the ability to just connect a keyboard to it. My own control panel has a keyboard (4x4). I plan to use it to program more advanced features like adjustable pulsing. That way I can do all adjustments using software and keyboard input, rather than numerous pots.
i



Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Fri, 17 Nov 2006 03:13:07 +0000 (UTC), Ignoramus25220

Microcomputers are used in lots of hostile environments. Thing is, getting them bulletproof often takes some development time. For automotive and military stuff, read weeks in the EMI lab. While it's "the right answer" for a product developer, hackers like we would rather just build something that works than make a big project of it. YMMV!



Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

I did not do much to protect it from EMI aside:
- it is in a different compartment than the HF unit (though the welder currently has its sides open)
- It does not switch any wires (with its internal relays) that go into that compartment. It switches relays that connect those wires.
When I put my inverter into my welder, which I have not done yet, I will not connect it to Cubloc in any way. The inverter will have its own timer.
i







Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
In years past - the DEC MicroVax workstations - I ran DEC-ELN which was a kernel OS and above a shell was only user code while below it was DEC's. It was MIL Spec.
The interesting thing, being (company was at the time) the world largest buyer of DEC machines - I was also supporting DEC as a vendor. The DEC UVax software team had no idea that the OS was even written for their machine (by a software group in a different division :-)).
But now days - one only has Sun software and Linux.
Martin
Martin H. Eastburn @ home at Lions' Lair with our computer lionslair at consolidated dot net NRA LOH & Endowment Member NRA Second Amendment Task Force Charter Founder IHMSA and NRA Metallic Silhouette maker & member http://lufkinced.com /
Don Foreman wrote:

-








Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

We should perhaps take this offline, though nobody's screaming yet. For DC-DC conversion to get negative bias for opamps, I've used the little chargepumps like the TC7660 -- 88 cents from Digi-Key. I've also done it with the venerable 555, a cap and a couple of diodes. Negative bias usually doesn't need to be regulated. There are also a lot of little switcher chips that can be set up to deliver negative output, and the little red toroidal 35-cent inductors (Hurricane electronics) at Ax-Man work nicely. They're about 220 uH, way more than necessary but I like the 35 cent part!
I think I would use an N-channel MOSFET with a high-side driver. That's partly because I have LOTs of N-channel MOSFETs in my goodiebox -- and a few IR high-side drivers as well. N-channel devices are cheaper, more readily available, and exhibit lower Rds (on) for given die size.
Some former colleagues, Zoltan Zansky and Norm Planer, once patented a rather clever high-side drive scheme that can work at very low frequencies. (The high-side driver chips rely on the MOSFET switching at a fairly high frequency.) Go to www.uspto.gov and search for patent # 4,443,719 This was in 1982 so the patent is expired by now and therefore in public domain.
I'll put a log-taper pot on my list for next Digi-Key order.
Gee, if a guy used a PIC or AVR then any old pot would work. <G> I'm just now getting started with PIC's and AVR's, but I am very familiar with the good ol' LM324! Probably have a dozen of 'em in the linear goodiebox. The LM324 was originally designed to be a rugged automotive/industrial part. It's been around for two decades or more and still is in full production by many vendors.
Use of a uP would enable including more "features", but it's good to learn with a basic and bulletproof device first. As Ernie has observed, there isn't much lore extant about pulsers. I wouldn't know what the hell to do with a CyberTig if I had one.








Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Nice project
Ironic though that the New miller equivalent (Synchrowave 200) has one built in now.
Or was this project more about doing it than the feature itself?
Ken Moffett wrote:






Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Brent,
That's a rabbit that I'm not going to try and chase. Miller seems to add new features as soon as you have purchased one of their machines. ;) The pulser was not a necessity, but seemed like an interesting project...with a potentiol practical outcome.
Ken




Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
I dont blame you for that one
I havnet used to pulser yet. I dont think my good old fashioned TIG welding is up to snuff yet doing plain old welding so i'm not going to get fancy till i can do plain and reliable
But out of curiosity what model and type of connector is that 14 pin one?
a finger amperage control for out of position work would be a nice project in my books (though i prefer the pedal) and the only part that i havent stumbled across (Or thats sitting in my electronics parts pile) is that Amphenol connector.
I know Amphenol makes them but i havent found it (too big a forest of products they make to find the 14 pin welding connector tree)
But Great work on the pulser unit
Ken Moffett wrote:









Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

one
;)
Bernt,
The connectors are listed in my TIG Pulser.txt file in thre Dropbox. They are:
The AMP/Tyco Connectors are available from Electronic Industries Inc. P.O. Box 266 Oshkosh WI 54903-0266 (920) 235-8930 http://www.electronicind.com
Part numbers: 1 each: AP-213571-2Plug 1 each: AP213570-1Receptacle 1 each: AP206070-1Plug Shell w/Cable Clamp 10 each: AP-213603-2Pin, 18-14, Tin, 10 each: AP-66601-2Socket, 18-14, Gold Total (w/ shipping) $27.82
There's no magic in these connectors. They only used 5 pins. As I said in another reply, if I had thought about it I could have cut the foot control cable near the machine end and inserted any 5-pin connector pair I wanted. Then use that as the connecting point for the pulser. Even without the pulser you could do that for your finger control. Were you looking at the finger control ("tig remote.xxx") files posted in the Dropbox? Looks pretty good.
Ken









Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Related Threads

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.