When welding with an AC arc welder, it is often necessary to “scratch start” them to get the arc going. For those unfamiliar, it is just like it sounds. You drag the head across something just like a giant match. There are some that come with an arc stabilizer or “high frequency starter”. This is preferred, but they can be hard to find. [Bill] shows us how to make one of our own. Though you may have an easy enough time finding a big transformer, you might run into some difficulty finding the capacitors, and tungsten spark gaps. If you manage to get your hands on them, you can follow [Bill]’s schematic and build one of these starters for yourself.
52 thoughts on “High Frequency Start Box”
Leave a Reply
Please be kind and respectful to help make the comments section excellent. (Comment Policy)
looks intressting but the parts are not easy to find ;)
when I was building my tesla coil I got some tungsten spark gaps and similar caps form ebay.
Did the tesla coil work? What are you powering with it?
My first thought an auto ignition coil will not stand up to continuous duty. Second thought was that my be why it’s labeled a start box not a HFHV arc stabilizer and what appears to be foot switch. Third though was #8 wire in pretty puny when compared to my welding cables. The only other feature I could think to add is current sensing, along with time delay. That would sense an attempt to start the arc, turn the box on and, turn it off when the welder had time to establish the arc. Allowing one to get rid of the foot switch
That is the basic idea behind a lift arc start welder. When you touch the tungsten to the workpiece, the welder senses a short circuit, than when you lift the tungsten back up the welder senses that the circuit is now open and starts a high frequency arc.
Very interesting. Only question I have is “why tungsten spark gaps?” Wouldn’t regular spark gaps work?
Why Tungsten? I would imagine Tungsten has an extremely high melting point and has a good resistance to arc oxidation. So they last a lot longer.
Electronics Goldmine sells 35kv capacitors like the one pictured.
Why spark gaps at all? Wouldn’t a good Power Semiconductor do the Work, the same kind they use in tesla coils.
The purpose of the spark gap is to isolate the HV supply from the 200 amps of current from your welder that would fry you HV power supply faster then building 7 went down if directly connected!
And to keep the voltage from getting to high and giving you a good zap
green coded tig welding electrodes are almost pure tungsten. they are available from welding suppliers in many different diameters. a spark gap can be made from a couple pieces of an electrode that are shimmed apart.
Great Hack !!
Spark gaps because its simple, spark gaps never blow up like a transistor. Im sure theres probably also some better reason related to the incredibly high frequency.
To isolate the HV supply from the 200 amps from the welder to avoid the magic smoke…lol
To be able to superimpose this high frequency signal on the secondary of the welding transformer, you also need a high-pass filter connected in paralel with the secondary winding.
Otherwise the high frequency signal will feed back into the primary.
Hi, I am very new to tig welding and have just got an inverter tig welder thats scratch start. my question is, do the keep the arc starter connected all the time while i am welding or is it just switched on to get the arc then switched off. thansk for help.
On steel, we’ve found that it can be switched off once the arc is started. From a very short test on aluminum, we’ve found it needs to be left on. There is probably either a known exception to this, or a skill level which we would need to pass so that it can be shut down, but so far we don’t know how.
see this schematic of the Miller HF-15 arc starter: http://www.yoreparo.com/foros/files/hf15.gif
Hi!
The arc starter works really good !
You dont need touch the stick electrode with the work to initiate the arc. Very usefull for low current weld.
Thanks to Willian F. Dudley Jr. that published the schematic PLEASE NOTE THATH THE 500 pF capacitor is TO BE CONNECTED IN SERIES with the primary of auto ignition coil. 100ohms not required im my assembly.
Also the auto ignition coil was replaced with a homemade transformer primay 10 turns #22awg, secondary 300 turns #28awg, core from a tv flyback “c” 200mm2 cross section, insulation between layers 0.12mm polyester paper. No more than 38 turns in each layer. Alkid (exterior spar vernish or sparlac)impregnation.
The termminal of buzz box to be shunted with 0.5mF x 400v cap to prevent HF voltage enter into the main weld machine
Enjoy
Newton
hello , i did the circuit and it works great on itself, but i burt my rectifier diodes on my inverter twice after few seconds,
i see here that you have to add capacitor, but i ‘d like to make sure about how.i understand that it should be 0.5 microfarad between work lead and electrode lead, am i right?
thank you so much in advance .
sorry for my english
jeremy
By the way, the 500pF x 30000V capacitor was replaced by a 1000pf x 20000V home made capacitor
Material required : aluminun foil used at kitchen and glasses from home windows.
Enjoy
Newton
Newton.b@live.com
How wide do you set the gaps on the tungsten? If its like a spark plug, its gonna need a certain size gap.
What voltage is supplying this start box?
what Amount HF Generator Voltage?
Yes, it works very fine. Connected to the buzzbox allows u weld very confortable at 30-80A.
The schematic has a mistake: 500uuF in parallel with the 10 turns primary.
To fix make the 500uuF cap in SERIAL with the primary 10 turns coil.
The 500uuf capacitor was made with windows glass e aluminiun cook foil.
The tesla coil was simplified too, using simple ferrit beats found in scraped computer monitor.
Assembled with 2 gaps only,
At all is very effective device to help weld at low currents.
Newton Brawn,
newton.b@live.com
Info gleaned from Miller HF-15 manual(TM-610A ,Sept 1997):
Can be used with other processes such as Plasma Arc Cutting! (hint hint)
Spark gap is .008 inch.
High voltage transformer steps up voltage from
115v to 3000v(three thousand).
High freq produced is approximately 1 to 2 megahertz.
please, send me de schematic of the high frequency start box. thanks
I know the post talking about the coil standing up to continuous duty is way old but I would like to point out that a car coil should do fine durability wise. On a V8 engine a conventional coil would fire around 400 times per second at 3000 rpm. 3000/60=50 full engine rotations per second. The coil fires once per spark plug. For a V8 that would be 8 firings per revolution so 50*8=400. An automotive coil has a service life of years and is cheap to get as well as easy to deal with.
Just wanted to point that out for anyone coming across this information.
A four stroke fires each cylinder once every two revolutions so 200 times per second at 3000 rpm and is powered by 12 volts not 115.
I recognise that this article (and posts) are very old,but can asnybody provide schematics for this project please.
Thanks Rob.
Schematic can be found here:
http://www.casano.com/projects/hfstart/index.html
No they can’t
Can someone build me one of these? I like to work with old vintage welding equipment and this would be a nice add-on to an old dinosaur stick machine I have.
what is the name ds parts to complete this assembly not understand that piece that are connected wires Green black and white I thank
I don’t see any responses to your question, so I will do so now. On the left side of the picture, you can see the back side of the IBM cord connector. The green wire is chassis ground, the white wire is your neutral conductor, and the black wire is your power in feed wire. When you plug an IBM power cord like you find on most computers, that will provide the electricity to operate the unit.
I have a Dayton 3AC01 high frequency arc stabilizer. What is it for???
Can anyone comment on why this uses 2 gaps rather than 3 or 1 gap? What is the impact on number?
Wouldn’t non-resistive motorcycle spark plugs do, to replace the expensive tungsten dork gaps?
You could grind off the ground tang and mount two, electrode to electrode. The threaded body would allow for gap adjustment.
They’re well insulated. They are heat resistant. And they cost about $2 each.
Just a thought.
*Can someone please verify that the schematic that has been linked to is accurate?
In my eyes, I see the 500pF cap in diagram as parallel.
*What kind of “lamp dimmer”? Seriously there are several.
*Can a flyback transformer be used instead of ignition coil?
*Someone posted, frequency of unit is in the Mhz… accurate?
Thanks a lot!
how to off arc mma trigger is off?
If you need (2) spark gaps, could (2) spark plugs both threaded into the same piece of metal with the high voltage entering the high voltage terminal of one spark plug and exiting the high voltage terminal of the second spark plug ? Just remember to isolate the metal that the spark plugs are threaded into from any surrounding metal to prevent you getting the high voltage ! ! !
Anybody have an active link for schematic?
See below internet archive reference: https://web.archive.org/web/20161127231205/http://www.casano.com/projects/hfstart/index.html
Can someone please repost the schematic
Looks like the internet archive has a post from Nov. 2016:
https://web.archive.org/web/20161127231205/http://www.casano.com/projects/hfstart/index.html
Today we setup and ran our build of the High Frequency Start Box, and it worked perfectly! Our shop foreman was thrilled to add it to his set of tools. The setting on the lamp switch dimmer setting seems to have limited impact on the results of the arc started, though we may learn otherwise with time and experience. He was able to lay down a nice bead in some aluminum test material with his Hobart stick welder attached to a tig torch and the HF box. The way the arc jumped and started immediately was incredible. It jumped between the test material and the tungsten at about 5/16″ of separation.
When working with steel, the HF start box helps get started easily, but can be turned off once the arc is started. In aluminum, leaving it on made the arc much more stable, and gave a broader warning zone showing that you are moving out of the target separation between electrode and work piece, without breaking the arc immediately. Seems like it would help learners develop a feel for proper positioning more quickly. No idea on what impact it would have on the life of the spark gaps or any other component of the system.
We used 118X80X20mm Ferrite Core… it was larger than necessary. If we build another we’ll try a 102x65x20mm or a 85x55x20mm Ferrite core.
An important note for anyone replicating this, be sure that the ground is properly connected to your mains ground… otherwise you end up with strange electrical feed backs. It caused an auto feed table on a mill in the shop to start feeding until we tracked down a faulty ground in the electrical outlet we had the HF box plugged into for testing. There was definitely some sort of undesirable feedback going on without the ground properly connected.
Hi Matt.
I just build a HF start box, and I am not happy with arc length.
Its only around 1/8”. Would you be able to help me to point a possible problem.
I took a picture of my setup and and I can email it to you.
Your help will be greatly appreciated.
Thanks.
Hi Alex,
I’ll be happy to help if I can.
Thanks Matt for your quick reply.
I build my box using components showing on schematic.
But HF arc is only around 1/8” not like yours which you
post is 5/16”. I would like to send you a picture of my
box but I need your email address. Maybe you change
something and that’s why it works better.
Thanks.
Please where can I find the schematic or direction on this setup ?? Thank you in advance
Look here: https://www.dudley.nu/projects/hfstart/