[Simclardy]

I hear alot about these modern inverter machines and their great features.
I thought i would start compiling actual images from my welders with the hope of eliminating some of the magic/mystery associated with the new technology.
All the companies talk about wave shape manipulation, but i never see actual images. Over time this thread might be able to answer questions, like, why is this brand producing such stable arcs? Or why is it so good at arc starts?
My background is electrical. I repair all manner of appliances and some motors but I am not an engineer. I am a hobby electronics enthusiast. So if there are professional electrical/electronic engineers in the crowd please jump in and help us.
If your not a PE but can add to the subject or correct me, please do. i am forever a student.

Ok. I should have snapped a photo of the settings. The top left shows time per division and bottom left shows voltage per division. I have the machine set to AC, 200hz, balance 30%ep, 125 amps. The first image (1ms/div) is the overview with max voltage about 100 volts. I zoomed in a bit to show the change of state from negative to positive, and the third image is zoomed in further to show the switching speed and stability of the voltage.

The welder works well for me but i have never used another tig welder in 5 years. I am in the market for a new dynasty, fronius, or htp.

I was surprised a bit by the final image but i can't say if it's good or bad until i compare. Just seems a bit rough. At 50,000hz it averages out to some effective voltage. For all i know it could be the perfect switching rate.
You can see that the rms for the positive is higher than the negative side (zero crossing is the yellow arrow on left). This machine does not have "asymmetric" wave control. Thought that was interesting.


Let me know if you want me to run a specific test and i will try.

Cheers

Sent from my SM-G930V using Tapatalk

[Poland308]

Looks like a square wave imitation.

[Simclardy]

Looks like a square wave imitation.

Yeah, square wave, is used loosely. This might be normal? I suspect a more expensive unit uses better electronics.
When i get a chance i will run the test again and use my clamp meter. This will give us a better idea how the current is operating.

Sent from my SM-G930V using Tapatalk

[Arno]

This might be normal? I suspect a more expensive unit uses better electronics.

Look up IGBT or MosFET 'ringing'. It basically overshoots and takes a little while to get to the target voltage.

Better or higher frequency driving electronics and some more smoothing circuits can turn this into cleaner square waves.

However.. For a welding machine (and not an amplifier or something like that..) the question quickly becomes if a 'cleaner' signal would work any better for the performacne welding arc than a 'dirty' one so if the added expense of improving the circuits is really worth it.

I can imagine that from a pure 'will it work' viewpoint a 'dirty' signal will work just as well. A better controlled 'clean' signal may give a smoother feel/response when wleding though. That's what you often notice between cheap-o and more 'brand' welders where the latter ones are more consistent and the weld arc experience is 'smoother' somehow.

Also when looking at welder signal outputs it's good to also use a current clamp as a lot of the manipulations in the 'signal' are not only voltage based but also current. (yeah.. TIG is 'constant current', but that's also easy to play with on an inverter..) As the current will be influenced/delayed by inductive effects in the system it's response may well be smoothed out compared to voltage spikes.

On youtube This Old Tony also did some scope testing on his old (and fairly broken ) TIG welder if you want to see some extra info.

Bye, Arno.

[BillE.Dee]

Is it the dirty signal that causes the stutter on start and lower amp settings on the "lower cost" machines ??

[Simclardy]

"However.. For a welding machine (and not an amplifier or something like that..) the question quickly becomes if a 'cleaner' signal would work any better for the performacne welding arc than a 'dirty' one so if the added expense of improving the circuits is really worth it."-arno

That is a question i want to be able to answer.

I do plan on running the test again with my clamp probe. MOSFET's and IGBT's are both used in switch mode power supplies (SMPS). By design they are "digital" and not linear. So i agree the effective amperage will be more interesting than the voltage.

Thanks for the lead on old Tony. I will try and look it up.
Cheers

Sent from my SM-G930V using Tapatalk

[Simclardy]

I can't answer the question......yet. just to be clear my images where taken after the arc was stable. The high frequency start is something i would like to investigate but did not want to complicate my first test. I was also worried that the high voltage could damage my equipment so i was trying to establish arc as soon as possible.
My probes are rated to 7kv+/- . The highest spike i saw was 1.5kv so i feel better going forward.
As for arc wandering at low amperage this could be due to sloppy switching aka inferior SMPS.
I am looking to purchase a dynasty 210dx or the fronius magicwave 230i. I am dying to know what the actual differences are in the power output. Unfortunately i don't have access to try a unit before purchase.
Cheers

Sent from my SM-G930V using Tapatalk

[Simclardy]

This might be normal? I suspect a more expensive unit uses better electronics.

Look up IGBT or MosFET 'ringing'. It basically overshoots and takes a little while to get to the target voltage.

Better or higher frequency driving electronics and some more smoothing circuits can turn this into cleaner square waves.

However.. For a welding machine (and not an amplifier or something like that..) the question quickly becomes if a 'cleaner' signal would work any better for the performacne welding arc than a 'dirty' one so if the added expense of improving the circuits is really worth it.

I can imagine that from a pure 'will it work' viewpoint a 'dirty' signal will work just as well. A better controlled 'clean' signal may give a smoother feel/response when wleding though. That's what you often notice between cheap-o and more 'brand' welders where the latter ones are more consistent and the weld arc experience is 'smoother' somehow.

Also when looking at welder signal outputs it's good to also use a current clamp as a lot of the manipulations in the 'signal' are not only voltage based but also current. (yeah.. TIG is 'constant current', but that's also easy to play with on an inverter..) As the current will be influenced/delayed by inductive effects in the system it's response may well be smoothed out compared to voltage spikes.

On youtube This Old Tony also did some scope testing on his old (and fairly broken ) TIG welder if you want to see some extra info.

Bye, Arno.

I did check out this old Tony. Good stuff. Can't wait to compair.

Sent from my SM-G930V using Tapatalk

[Simclardy]

Ok. The yellow is voltage and current is blue. 1millivolt equals 1amp. So i have the machine set to about 35 amps. If you had a clamp meter you would say great. The rms is 39 amps. Frequency setting is aslo pretty accurate. You can see 500microseconds per division which works out to be 200hz very close to rotary setting. So the bad. I could not get below 10-11 amps and even at that it would randomly fire the high frequency start. Not good when you are trying to baby an edge. I need to call everlast because the literature claims 5a +/- 2a.
Maybe this is only for lift arc?
The other issue is, the wave looks awful. This old Tony's wave looked better except his was inconsistent. My wave is consistently ugly! I prefer ugly but consistent.

I will take some shots at slower hz and see if it always overshoots this bad. So like i said. A standard clamp meter would show the 35-40 amps. But the scope is showing 65 amps on electrode negative!

I think i will try and capture at minimum amps.
This explains a bit to me.
Don't get me wrong the welder has made some nice welds but i think i am in for a nice surprise when i finally upgrade.


Sent from my SM-G930V using Tapatalk

[Simclardy]

I just performed the test at it's lowest setting. The rms is 13.3amps and the max EN is 16amps. The third photo shows the voltage spiking over 5000 volts!

I have to say this welder is pretty rough at the low amp setting

Sent from my SM-G930V using Tapatalk

[Oscar]

Interesting stuff. I wonder how other welders stack up.

[Simclardy]

Interesting stuff. I wonder how other welders stack up.

I just ordered a brand new dynasty 210dx. So we shall see the difference soon enough.

Sent from my SM-G930V using Tapatalk

[Oscar]

Interesting stuff. I wonder how other welders stack up.

I just ordered a brand new open box dynasty 210dx. So we shall see the difference soon enough.

Sent from my SM-G930V using Tapatalk

Nice. I'm looking forward to the measurements.

[Simclardy]

This is the dynasty 210dx.
200hz
75% balance
30amps
Advanced wave
Blue line is amps
Yellow volts

https://uploads.tapatalk-cdn.com/202001 ... 19d10d.jpg[/img]

Sent from my SM-G930V using Tapatalk

[Simclardy]

This is the dynasty 210dx.
200hz
75% balance
30amps
Soft wave
Blue line is amps
Yellow volts

Sent from my SM-G930V using Tapatalk

[Simclardy]

This is the dynasty 210dx.
200hz
75% balance
30amps
Sine wave
Blue line is amps
Yellow volts

Sent from my SM-G930V using Tapatalk

[Simclardy]

This is the dynasty 210dx.
200hz
75% balance
30ams
Triangle wave
Blue line is amps
Yellow volts

Sent from my SM-G930V using Tapatalk

[Simclardy]

So my observations:
1. The cleaning action (positive side) is not effected by the wave selection
2. The amperage does not overshoot as bad as the everlast
3. The reaction time of the scr or igbt, whatever miller uses is almost twice as fast as the everlast and appears to be much more stable or consistent (compared photo below to everlast image to see the frequency of oscillations)
4. The beads i stacked are the first 3 out of the box. Too excited to clean or get a smooth surface to slide so i just twisted my wrist. Top was first with 75 balance, bottom 2 are 80% balance.
I thought it was smoother than the everlast.
I have not pushed the low amperage limits yet and that is where i expect the biggest difference.
5. The stick welding was noticeably smooth compared to what i have been using.


Conclusion: i like to buy quality tools but i don't like paying for a name brand.
If the name brand happens to delivers a better product in an ethical way i consider the extra cost worth it. I see the R&D that miller has published on the web along with lincoln and i notice. Every time i call miller with technical questions i get a very knowledgeable tech within minutes, i notice. I have used some of Miller's products and i notice a difference.
I purchased the wireless foot pedal for $700 and and thought about the expansion card for $500 and damn it miller i notice. Lol

I will just add that i don't trust bean counters and i have seen many good companies erode their quality and ethics for profitability. Miller is not immune from temptation, so i hope they never become just a "name"
Cheers and happy new year!

Sent from my SM-G930V using Tapatalk

[Coldman]

Thanks for carrying out this investigation, really interesting. It confirms what I already knew but could not understand why. I have no experience with Everlast or lots of other brands on the market, but I have used quite a variety in the past and also owned a few that I moved on because I did not like the arc (weirdness thereof). Nothing came close to my treasured Dynasty 200DX. Once you use it you don't go back down to lesser machines joyfully.
I too have the wireless peddle and won't go back to a wired peddle again. Only blue for me. As long as they don't just become a name in the future as per your comments.

It actually makes a difference to me, I had more xray drops with the machines I moved on. If I drop a rare xray with the dynasty it always has an explanation like parent metallurgy faults, contamination or plain old laziness. Never left scratching my head wondering why because I did everything right.

[Simclardy]

Thanks for carrying out this investigation, really interesting. It confirms what I already knew but could not understand why. I have no experience with Everlast or lots of other brands on the market, but I have used quite a variety in the past and also owned a few that I moved on because I did not like the arc (weirdness thereof). Nothing came close to my treasured Dynasty 200DX. Once you use it you don't go back down to lesser machines joyfully.
I too have the wireless peddle and won't go back to a wired peddle again. Only blue for me. As long as they don't just become a name in the future as per your comments.

It actually makes a difference to me, I had more xray drops with the machines I moved on. If I drop a rare xray with the dynasty it always has an explanation like parent metallurgy faults, contamination or plain old laziness. Never left scratching my head wondering why because I did everything right.

No problem, thanks.
I have been runing some beads and i think the etch zone is much tighter. I will have to set them up side by side which is a pain but i just might.

The testing was not purely for the experiment but it provides a nice base line incase something goes wrong i can compare.
Cheers


Sent from my SM-G930V using Tapatalk

[Tabs123]

Hi Simclardy, Thank you for all this info it is much appreciated . I have a 200amp ac/dc tig welder that is very popular here in Australia (for a welder thats under $1800 aud) It has a digital screen and balance is just +10 to -10% and the user manual and manufacturer does not know at all what this means in regards to EP or EN percentages. So I am wanting to buy a current clamp and also a PC based oscilloscope. Could you tell me if I will need the Oscilloscope to have AC coupling mode? As the models Ive been looking at only are Dc coupling. Will a AC output on a Tig welder have any DC in it that will offset the signal being read on the oscilloscope. As I would not mind also checking my welders amount of EP and EN amplitude in regards to 0 and see if they are even in amplitude. Sorry im a bit new to oscilloscopes etc. Cheers from Australia. Regards Paul

[sbaker56]

I've actually been thinking about your results a lot lately and glad it got bumped and I could find it. I would've suspected Miller and Lincoln were importing a lot of the same or similar components from China that Everlast and other companies used and assembling it here with a massive mark up. I figured there was probably higher quality control and better build quality, but the electronics themselves would be the same.


After seeing your results I wouldn't be surprised if Lincoln and Miller might actually use some domestic electronics or at least spec higher end parts from China. It influenced me to pick up a used Invertec rather than an Everlast or other import around a month ago

[Simclardy]

Hi Simclardy, Thank you for all this info it is much appreciated . I have a 200amp ac/dc tig welder that is very popular here in Australia (for a welder thats under $1800 aud) It has a digital screen and balance is just +10 to -10% and the user manual and manufacturer does not know at all what this means in regards to EP or EN percentages. So I am wanting to buy a current clamp and also a PC based oscilloscope. Could you tell me if I will need the Oscilloscope to have AC coupling mode? As the models Ive been looking at only are Dc coupling. Will a AC output on a Tig welder have any DC in it that will offset the signal being read on the oscilloscope. As I would not mind also checking my welders amount of EP and EN amplitude in regards to 0 and see if they are even in amplitude. Sorry im a bit new to oscilloscopes etc. Cheers from Australia. Regards Paul

sorry for the late reply. i have been swamped with work since covid19 hit.
ac coupling just eliminates the dc element of the signal so that it centers the sine wave on the zero crossing.
there was a guy in welding web insisting on using an analog scope. i called Rigol and friends and i do not agree that you need the analog scope but they are cheap on ebay and would provide a good image for this type of signal.
cheers

Sent from my SM-G930V using Tapatalk

[Simclardy]

I've actually been thinking about your results a lot lately and glad it got bumped and I could find it. I would've suspected Miller and Lincoln were importing a lot of the same or similar components from China that Everlast and other companies used and assembling it here with a massive mark up. I figured there was probably higher quality control and better build quality, but the electronics themselves would be the same.


After seeing your results I wouldn't be surprised if Lincoln and Miller might actually use some domestic electronics or at least spec higher end parts from China. It influenced me to pick up a used Invertec rather than an Everlast or other import around a month ago

glad it helped. i would love it if folks added images to the thread. i feel like there is alot of rhetoric around which unit to buy, and not enough data.
cheers

Sent from my SM-G930V using Tapatalk

[xland]

HI ACE,
Nice thread on scope and nice beads.
Was wondering where you connected onto welder as I need to do this to continue my repair on my 200AMP ac/dc/PULSE Digital TIG.

alupulse 200D.jpg (61.55 KiB) Viewed 21476 times

2 IGBTs let go and a gate resistor ,also both bridge rectifiers.
Got it going did a bit of welding (DC SS 80 amp pulse)all good.
Next time I switch unit on circuit breaker blows..............

My forum intro link is here

https://forum.weldingtipsandtricks.com/ ... 27&t=15923

Would like to start a thread on repairing/testing inverter welders, also your input would be valuable.
I can make the thread public or PM you for advice if thats OK

regards
X