Arc welder triac controler

[oldway]

40V is the minimum open circuit voltage of the welding transformer to ignite the arc.
Commercial low cost units have an open circuit voltage of 45 or 50V.
It would be better to have an open circuit voltage between 60 to 70V.

[tautech]

40V is the minimum open circuit voltage of the welding transformer to ignite the arc.
Commercial low cost units have an open circuit voltage of 45 or 50V.
It would be better to have an open circuit voltage between 60 to 70V.

+1
But at the risk of minor shocks for the user. (wet work area, use outdoors, etc)

Been here myself.

[oldway]

I would recommand to rodrigopires to not waste his time with this kind of project.

I have some experience with MMA welding, both with welding transformer and DC welding inverter...there is simply no match at all....DC welding inverter is 1000% better than welding transformer.

Lets try welding with a welding inverter and you will understand what I mean.

There are now cheap welding inverters available, I recommand to buy one.
I have a MOS100 DECA inverter and I am very satisfied with it for hobby welding.

[eneuro]

It would be better to have an open circuit voltage between 60 to 70V.

I think, that I will try add in series with my welder transformer a few turns on this MOT core, which has broken crappy aluminium primary coil:



Than I will make the same amount of turns with provided HV low current added MOT core primary at much higher frequency than 50Hz using one of those step-up DC-DC converter with high voltage gain:


In the case of AC welding, since I know current direction from my Hall efect sensor, I can step-up voltage in series with welder transformer synchronized in polarity in a way that this HV low current MOT voltage will create sparks between welding elctrode and base metal regardless of transformer low voltage 

What do you think, If I made usch plasma speaker talking during welding to take care and do not touch welded metals and electrode it has enougth ignited arc to start welding easy, providing current will be vey low and kept on safe levels even if human touched this think-only small shock to remeber to do not touch this in this welder? 
Plasma Speaker - Singing Arc: HV ZVS Flyback driver project


So, question is how much ignition arc energy (we can break air at given distance with different current) is needed to start welding easy, because probably it matters, while HV in series with welding electrone and metal can be made in many ways.

I like this idea with additional MOT core and I'd like to test this while I've low voltage bulky transformer (its secondary is now used with a few thick turns of wire as spot welder) , but Thinking if it could be used as classic welder too, since its oryginal secondary has quite thick 2 wires in parallel of copper...

Those copper pipes used to help dispate heat on spot welder secondary are 15mm in diameter on this photo, to get some idea of trafo dimensions 

Idea is add afew lones of code to support not only spot welder, but add classic welder mode too with not removed trafo oryginal secondary 

[rodrigopires]

Rodrigopires did not explain what transformer he is using.
I hope he knows that the welding transformer must have a high short circuit voltage (20%) to limit short circuit current and an open voltage above 40V to work correctly.

I'm using MOT's, i used two, but now with the power controler (not yet finished) i'm going to ad a third one. With two MOT's i had about 30V and it welded about anything, the only limitation was my home circuit breakers .

I would recommand to rodrigopires to not waste his time with this kind of project.

I have some experience with MMA welding, both with welding transformer and DC welding inverter...there is simply no match at all....DC welding inverter is 1000% better than welding transformer.

Lets try welding with a welding inverter and you will understand what I mean.

There are now cheap welding inverters available, I recommand to buy one.
I have a MOS100 DECA inverter and I am very satisfied with it for hobby welding.

You are right, there are some cheap and good welders out there, and with the money i have already spent on this project i would be able to by one, but whats the fun? Why buy one when i can make one, yes it not as good, but its homemade.

[eneuro]

With two MOT's i had about 30V and it welded about anything, the only limitation was my home circuit breakers .

That is why I prefere active monitoring of welder transformer primary AC current, so using Hall effect current sensor we can calculate average current drawn in lets say 100ms periods, so when it is higher than given percentage of home circuit breaker triac/thyristors fire angle can be adjusted in both half AC periods to avoid overloading home grid 

Simple circuits as yours might work, but blackout at home while welding for fun this is not what circuit designer wants to happen, even in hobby project....

[oldway]

MOT transformers are ok because the windings are wound side by side, which give a high short circuit voltage.
15V by secondary seems reasonable given the power rating of the MOT's.

So yes, it takes 3 MOT's with the secondaries in series and primaries in parallel.

The short circuit current is limited naturally by the short circuit voltage: 20%, this means that the short circuit current is equal to 5x the rated current.

You should know what open your home Circuit breaker: inrush current or short circuit current ?

Regarding to the primary phase control, it can not be fast because a rapid change in phase angle produces a DC component that saturates the transformer and produces a huge current in the primary.
The cure can be worse than the problem it is trying to eliminate.

This is one of the major limitations of the phase control by the primary voltage of the transformers.
Whenever possible, I recommend to always do the control in the secondary than in the primary.

As I said, this circuit is very simple but also very imperfect.
For small transformers, it is acceptable.
I have used it for the project of a small battery charger with 12V 8A ajustable output current and charging voltage limited to 14.4V. (see pictures)
But for higher power, you should use a more sophisticated circuit.

[Circlotron]

As I said, this circuit is very simple but also very imperfect.
For small transformers, it is acceptable.
<snip>
But for higher power, you should use a more sophisticated circuit.

You'll love this then. In the late 80's I worked for a company that made SCR based UPSs up to 20KVA, and over the road was the electric furnace division. The used to make their own transformers from scratch too. Anyway, there was this furnace that was IIRC about 20KW, maybe even a bit more, and there was a 1:1 isolating transformer between the incoming 3x240VAC and the heating elements. The tranny had a star primary with the star point unconnected, and a delta secondary. The phase control part was 3 x SCRs and 3 x anti-parallel diodes. Most definitely the two-stroke topology of the phase control world. At some settings that transformer made this dirty, sickening growl. My protests went unheeded.

[oldway]

This is a modification I made in 1993 of a 6000A 12V water cooled electrolysis rectifier which was split in two 3000A 12V rectifiers each with his own control.
I made the new 3 phase control circuits and on site modifications.
Phase control was made in the primary of the transformer by a 6 SCR's.

[eneuro]

Regarding to the primary phase control, it can not be fast because a rapid change in phase angle produces a DC component that saturates the transformer and produces a huge current in the primary.

Thanks for this hint.
I'm talking about 1ms adjustements, so while @ 50Hz we have 10ms half AC mains period in 10 steps can be reduced to small currents 

Anyway, while I've AC mains voltage zero crossing based phase shift timing possible, but I've also Hall current sensor like this below and MPU makes 10000 current senses per second, so I know not only when voltage zero crossing happends, but also when current goes to zero



Trying to optimize my phase shift controll and I've used so far voltage zero crossing to detect positive AC mains pulses, take measurements within 2 seconds to find exact period time and calculate needed phase shifts to fire two SCRs in anti-paralle shown earlier.

But, what about using current sensor? I've 60W light bulb on SCRs switch and the same 60W in parallel with transformer primary, so even when SCRs switch is OFF this 60W light bullb in series with welder/spotwelder transformer, limits circuit current, but I can detect current maximum, so period, than expected zero crossing, etc...

Maybe I could  take advantage somehow of fusion data from voltage zero crossing and current sensing (crossing) when SCRs switch is off but, small Ac current present in circuit, while 60W light bulb bypasses SCRs switch? 
I've upgraded my Hall effect sensor for higher current sensitivity and in th ecase of spot welder I'd like to try detect for example rise in current on traf primary, when spot welder electrodes are closed on welded steel, so within 1s-2s activate choosen spot welding program without need to press any spot start buttons etc.

Maybe I will try the same with classic welding-touching electrode will close welder transformer secondary )increase primary current and heat up welding start point) and withiin 1s-2s welder transformer will start increase its power and HV circuit activated to help ignite arc....   

I need experiment more with this enhanced Hall current sensor, if I will be able detect closed welder transformer secondary using DSP processing in MPU with 10kHz current sampling rate 

[rodrigopires]

With two MOT's i had about 30V and it welded about anything, the only limitation was my home circuit breakers .

That is why I prefere active monitoring of welder transformer primary AC current, so using Hall effect current sensor we can calculate average current drawn in lets say 100ms periods, so when it is higher than given percentage of home circuit breaker triac/thyristors fire angle can be adjusted in both half AC periods to avoid overloading home grid 

Simple circuits as yours might work, but blackout at home while welding for fun this is not what circuit designer wants to happen, even in hobby project....

that happened with my old setup, just two MOT's with secondaries in series and primaries in parallel straight to the wall socket with no protection or regulation,thats why the only limitation was my home circuit breaker. Thats why im building this regulator, to eliminate that problem.

[eneuro]

Thats why im building this regulator, to eliminate that problem.

But this simple regulator circuit has no current sensor, so still you have no idea what average and maximum current is, so even if welder transformers will not triger circuit breaker, someone will turn on microwave oven during your welding and.... blackout again 

But ok, for serious welding no way three phase additional grid needed, to protect single phase 230VAC home grid from overcurrent blackouts when playing with high power tools 

[oldway]

that happened with my old setup, just two MOT's with secondaries in series and primaries in parallel straight to the wall socket with no protection or regulation,thats why the only limitation was my home circuit breaker. Thats why im building this regulator, to eliminate that problem.

You still did not answered my question: when does the circuit breaker trip ?
At power on ? (this means inrush current)
Or when you short circuits output of the 2 MOT's ?
With 3 MOT's, things will be worse.
Perhaps are your home circuit breakers too low intensity for this purpose.
With commercial 150A welding transformers and 230V mains voltage, you need at least 25A circuit breakers.
These commercial transformers have an optimum transformation ratio, that's not the case of yours MOT's.

This phase angle control was intended for adjusting welding current (for exemple, from 50 to 150A), not for eliminate circuit breaker trip.

[oldway]

Thats why im building this regulator, to eliminate that problem.

But this simple regulator circuit has no current sensor, so still you have no idea what average and maximum current is, so even if welder transformers will not triger circuit breaker, someone will turn on microwave oven during your welding and.... blackout again 

But ok, for serious welding no way three phase additional grid needed, to protect single phase 230VAC home grid from overcurrent blackouts when playing with high power tools 

As I already explained, current regulation can only be slow with phase control in the primary.
It is not able to limit instant peak current. Circuit breaker is faster than phase control current limitation and it is not possible to avoid circuit breaker tripping by this way.
No need of a current sense, it is useless.

NB: The conventional solution of magnetic shunt welding current setting is far better than the phase control by triac.

[eneuro]

It is not able to limit instant peak current.
...
No need of a current sense, it is useless.

Nope, current sense is needed, and no need to limit transient current, but I've in this calibrated custom Hall sensor current limit set to 16A (avaregae of 10ms 10kHz sampling moving average-half period of 50Hz mains) on welder transformer primary which at 230VAc means 3.7kW, so it will not protect against short circuit, but when welding current will be inceased by potentiometer, safe level phase shift can be easy set, while this thing will turn off in current zero crossing for a few seconds when such overpower event happends.
Without current sensor you have no idea what this phase shift limit can be...until you make blackout in house, trying to find this point with potentiometer 



Another story is adaptive phase shift adjustments to rising input power (current). I will test those 1ms time phase shifts adjustments, hoping that below current limit set to 20A per 10ms half 50Hz period it will not create horrible waveforms, due to this phase shift fine tunning 

Update: Tested overcurrent protection software set to 3A with this 21W light bulb powered from DC 12V, so multimeter showd 1.63A current draw.

However, when 10ms averaging perido was set.... yep inruch current when cold light bulb was turned on trigered this current threshold of couse 
Than changed this time to ~0.1s (1000 averaged current samples) and it was enougth time to warm up bulb filnament to the point, average current sampled at 10kHz within 0.1s from cold start was below 3A overvoltage protection 
later, I've connected 55W car lights bulb and of course overcurrent worked all the time, since multimeter showed ~4.44 A current draw 

This experiment showed, that probably 0.1s might be fine during welding too, since I'll add also soft start (it will be used in spot welder in the same trafo) and maybe demagnetization of trafo core (in spotwelder where we have short circuit on secondary from th ebegining to the end).
Classic welding is slightly different, but I think it will be nice have primary nominal current set potentiometer, and in software automatically overwoltage current will be set based on those settings as lets say 25% higher-it turns off wedling for 10 seconds when something like this happends-enougth time to cut glued weld electrode from steel plate 

Now, calibrated this my new Hall effect sensor and upgrading software to implement this adaptive phase shift controll-overcurrent (overpower)  protection is already done and testing this right now, while preparing to myth buster of crimped soldered connectors vs only crimped   

BTW: Mag amp on this welder transformer primary in on my todo list-I have to upgrade MOT iron core with two windings and add thin contoll coil in the middle-It will be interesting to see if it works   

[Circlotron]

I'll bet that the power factor your transformer presents to the mains is low while actually welding because of harmonics from the triac phase control circuitry and lagging current because of (intentionally) poor coupling between primary and secondary of a MOT. Put a scope on the primary voltage and current and see if you can pull the current into line with a few power factor caps. Get yourself a few extra free amps before you trip the breaker.

[oldway]

so it will not protect against short circuit, but when welding current will be inceased by potentiometer,

It is during the short circuit that the circuit breaker will trip because short circuit current is only limited by the stray inductance of the transformer, that's more or less 5 times the nominal secundary current.
Yourself prove that current limiting is useless because it will not protect against short circuit and short circuit occurs in MMA to start the arc.

Why choose a Hall sensor?
With alternating current, it is better to use a current transformer, it is simpler, more reliable and more stable. (no drift, no need of power supply, ...)

[eneuro]

With alternating current, it is better to use a current transformer, it is simpler, more reliable and more stable.

Could you show us your simply CT circuit schematics connected to MPU pin?

I've shown in photo above: I take +5Vcc and GND from MPU and output from radiometric Hall effect sensor connect directly to MPU ADC pin with a few nF caps if I want additional filtering, but software moveing average at 10kHz sampling rate does the job in this (spot)welder MPu controller 

BTW: Forget about make CT at home, while with manufactured PCB my Hall current sensor PCB need to be only soldered one part: Hall effect sensor and ferrite core to guide mgnetic flux througth sensor....

Now, I'm working on improoved version, where opamps will be added, so with the same footprint shown above 25mm x 25mm this thing will be able drive additionally half bridge mosfets optoisolated gates in high power low voltage (safe 36V) many phase (three at least) synchronous rectifier drivers 

[oldway]

Could you show us your simply CT circuit schematics connected to MPU pin?

Do you really need a circuit schematics of a CT, a bridge rectifier and a load resistance with // filter capacitor ? You are kidding ? 

Forget about make CT at home

Are you kidding again ? 
Every 10 to 25VA transformer, without secondary (only the 230V primary), is a low cost CT.
You pass one turn in the free space and you only have to ajust the load resistance to have the right ratio current/voltage.

[eneuro]

Every 10 to 25VA transformer, without secondary (only the 230V primary), is a low cost CT.

It is useless-there is not too much space eg. to put 6mm in diameter copper wire 
This is quite nice, but costs.. 10x more than my <$3 Hall effect based current sensor with 10mm hole 

http://www.digikey.com/product-detail/en/C-CT-6/563-1390-ND/4428778


Anyway, BE CAREFULL when playing with CTs-DEADLY high voltages can be present on those not shorted CT output wires !!!!

Why the CT secondary should not be kept open?

I've choosen custom made Hall effect based current sensors for my high power EV subsystems and it works...

[oldway]

It is useless-there is not too much space eg. to put 6mm in diameter copper wire 

And what about toroïdal transformers ?
They are cheap, secundary can easily be removed and there is a lot of space.

In power electronics, you must choose the most reliable solution.

Here, we are not speaking about a project to be produced in thousands of units, only about a prototype of a welding transformer made with modified MOT's with phase control by triac.

That a secundary of a CT can't be kept open, that's basic electricity you learn at the very first year of study.
If somebody don't know this, better he does not deal with power electronics.

@rodrigopires: Be aware that the potentiometer is at mains voltage. For security reasons, it must be a plastic potentiometer (not only a plastic shaft) as PIHER type PC16 (http://www.conrad.fr/ce/fr/product/445663/Potentiometre-rotatif-Piher-PC16SH-10IP06474A2020MTA-Mono-02-W-470-k-1-pcs) and, as you can see on the pictures of my battery lader project, it is also mounted on an isolated epoxy board.

[rodrigopires]

You still did not answered my question: when does the circuit breaker trip ?
At power on ? (this means inrush current)
Or when you short circuits output of the 2 MOT's ?
With 3 MOT's, things will be worse.
[/quote]

The circuit breaker tripped when i was welding, after a while when the transformers got a bit warm, but it was very rarely, it did not happen in most of the time, just when i'm welding for a long time without stops, to cool the transformers down. And i hoped that with this regulator i would be able to reduce the heating on the transformers. An i chose to add a third transformer to reduce the load on the other two.

[/quote]
@rodrigopires: Be aware that the potentiometer is at mains voltage. For security reasons, it must be a plastic potentiometer (not only a plastic shaft) as PIHER type PC16 (http://www.conrad.fr/ce/fr/product/445663/Potentiometre-rotatif-Piher-PC16SH-10IP06474A2020MTA-Mono-02-W-470-k-1-pcs) and, as you can see on the pictures of my battery lader project, it is also mounted on an isolated epoxy board.
[/quote]

Yes, actually the pots i got are exactly like those, and i also got them in a separate board.

[oldway]

An i chose to add a third transformer to reduce the load on the other two.

That's not a good option: you will change de turns ratio and increase the mains current for the same welding current.
The circuit breakers will trip more often.
For the highest welding current without tripping the circuit breaker, the best choice is a welding inverter with active PFC (power factor correction).

Try to answer these questions and you will understand that what you try to do is not possible:
1) What's the minimun no load voltage of a welding transformer ? (NB: see output voltage of commercial units)
2) How much max welding current you need ?
3) Calculate the transformer power rating (even with 60% working factor)
4) How much mains current for max welding current ? Compare with your mains circuit breaker.
5) What's the nominal power rating of a MOT ?
6) How much MOT's you need ? (for the same power rating as 3))

[eneuro]

The circuit breaker tripped when i was welding, after a while when the transformers got a bit warm, but it was very rarely, it did not happen in most of the time, just when i'm welding for a long time without stops, to cool the transformers down.

I was thinking also on some tricky, magnetic shunt to add to MOT or even my E core type transformer to... add additional magnetic flux path at the top of MOT/trafo used in welding
I mean, put another mot on top of other to have its iron cores magnetic  contact and create addidional magnetic flux paths, so it should lower down flux in oryginal MOT secondary and current, but didn't tested it yet 
I'm thinking of adding steel sheet on the top, hoping it will decrease magnetic flux in oryginal transformer, so we could have output current regulation without  oryginal MOT/trafo core...
It will be interesting to see if it works, but I noticed, when turned last time my spot welder  that iron screw was heavy attracted to transformer core, so maybe it could work.

BTW: Of course be carefull to do not touch mains part of MOT reversed, when you put it on top of another one, since I expect HV there and maybe dangereus voltages...we'll see if it works in th ecase of my welder trafo, but magnetic circuit theory says, maybe? 

[oldway]

If you want to reduce output current (increasing the short circuit voltage of the transformer), why not use simply the original magnetic shunts of the MOT ?

 @rodrigopires: About overheating of the transformers, use forced cooling.
See how the welding current setting is done in the inexpensive welding transformer by adjusting the position of the magnetic shunt.