TIG training

38 views
Skip to first unread message

Matt Rutter

unread,
Oct 8, 2019, 7:31:52 AM10/8/19
to London HackSpace Metal Working Group
Hi all...  Is there anyone who'd be able to spare the time and give me a quick lesson on the TIG welder (paid of course).

I'm a seasoned MIG welder, but want to weld some aluminium for a project.... 

Matt

Des Quilty

unread,
Oct 8, 2019, 7:48:06 AM10/8/19
to London HackSpace Metal Working Group
Hi, I could help, but won't be around before the weekend of 19th Oct. Does that work?

If you can post some details of the project that would be helpful.

Matt Rutter

unread,
Oct 8, 2019, 8:01:26 AM10/8/19
to london-hackspace-m...@googlegroups.com
Hi Des.. Thanks!

There's no rush, as I don't have the metal yet, but can order it today.

I need to join two pieces of channel together to make some box, but only over a short length so it should be possible to weld inside too.  The channel is 80 x 40 x 5mm wall thickness.. 6060 grade. 

Matt

On Tue, Oct 8, 2019 at 12:48 PM Des Quilty <des.q...@gmail.com> wrote:
Hi, I could help, but won't be around before the weekend of 19th Oct. Does that work?

If you can post some details of the project that would be helpful.

--
https://groups.google.com/forum/?hl=en-GB#!forum/london-hackspace-metal-working-group
---
You received this message because you are subscribed to the Google Groups "London HackSpace Metal Working Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to london-hackspace-metal-w...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/london-hackspace-metal-working-group/e805bef1-c5a4-44ff-aac0-aee400a266ab%40googlegroups.com.

Jonathan Hodgson

unread,
Oct 8, 2019, 8:56:32 AM10/8/19
to London HackSpace Metal Working Group
Does anyone know if the TIG vs ACNode issue was fixed?

Tim Jacobs

unread,
Oct 8, 2019, 7:11:46 PM10/8/19
to london-hackspace-m...@googlegroups.com

It is still a problem, but only when using the HF overlay continuously (only for aluminium). The doorbot nodes have been hardened but the problem was still happening last time I tried it. I think there are further improvements to the nodes planned.

--
https://groups.google.com/forum/?hl=en-GB#!forum/london-hackspace-metal-working-group
---
You received this message because you are subscribed to the Google Groups "London HackSpace Metal Working Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to london-hackspace-metal-w...@googlegroups.com.

Des Quilty

unread,
Oct 8, 2019, 7:40:18 PM10/8/19
to London HackSpace Metal Working Group
I wonder if a shorter cable to the torch might help?

Dean also suggested putting the Tig on a different set of phases to what we have now, perhaps we are using the same phase as the 240 v supply or something along those lines

Dean Forbes (PERS)

unread,
Oct 12, 2019, 11:46:00 AM10/12/19
to London HackSpace Metal Working Group
did I .... that has its flaws too but could work 

I thought my best suggestion was to mute the so call door bell by a debounce routine that lasts for say 30-60 minutes - and as it is just the stalker on IRC watching it, is it a show stopper anyway - oops did I say that 

Dean Forbes (PERS)

unread,
Oct 12, 2019, 11:46:36 AM10/12/19
to London HackSpace Metal Working Group
did I .... that has its flaws too but could work - I mean is worth a try 

Jonathan Hodgson

unread,
Oct 12, 2019, 1:18:09 PM10/12/19
to London HackSpace Metal Working Group
I'm less worried by the door bell than by the fact that it was preventing doors from opening and also causing the Lathe to cut out (and I'm going to guess the mill and laser cutter might also have had issues if they'd been running at the time).

Dean Forbes (PERS)

unread,
Oct 12, 2019, 4:16:46 PM10/12/19
to London HackSpace Metal Working Group
you could wait for a fix but in the meantime I would go with a work around ...... 

Des Quilty

unread,
Oct 17, 2019, 5:03:03 PM10/17/19
to London HackSpace Metal Working Group
Matt, are you going to be around this Saturday?

Matt Rutter

unread,
Oct 17, 2019, 5:07:18 PM10/17/19
to london-hackspace-m...@googlegroups.com
Yes, I'll be there all day...I have practice material (and the parts I want to weld), plus new tungstens, and filler rods..... 
I fired the machine up today, just to try and get familiar with it... Was somewhat perturbed to hear the gas flowing all the time... isn't there a solenoid valve to cut it off when not welding?

--
https://groups.google.com/forum/?hl=en-GB#!forum/london-hackspace-metal-working-group
---
You received this message because you are subscribed to the Google Groups "London HackSpace Metal Working Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to london-hackspace-metal-w...@googlegroups.com.

Des Quilty

unread,
Oct 17, 2019, 5:15:33 PM10/17/19
to london-hackspace-m...@googlegroups.com
You need to adjust the post flow, the times on the front don't seem to match the actual post flow times. Turn it right down and the flow should stop straight away, then just turn it clockwise until you get about 10 secs post flow which should be enough to allow the electrode to cool sufficiently...

Matt Rutter

unread,
Oct 17, 2019, 5:50:10 PM10/17/19
to london-hackspace-m...@googlegroups.com
Ok.. that makes sense, except gas was flowing even before I turned the welder on. Hopefully it'll all make sense on Saturday....

Des Quilty

unread,
Oct 17, 2019, 6:24:45 PM10/17/19
to london-hackspace-m...@googlegroups.com
the post flow starts when you turn the welder main switch on for the first time. 

Des Quilty

unread,
Oct 20, 2019, 9:58:48 AM10/20/19
to London HackSpace Metal Working Group
Just looking through google for some help troubleshooting the TIG.

This is interesting...



Matt Rutter

unread,
Oct 20, 2019, 10:15:35 AM10/20/19
to london-hackspace-m...@googlegroups.com
Hi Des... That's the cap I thought might be the issue.... As mentioned in that thread they're not easy to come by.....

--
https://groups.google.com/forum/?hl=en-GB#!forum/london-hackspace-metal-working-group
---
You received this message because you are subscribed to the Google Groups "London HackSpace Metal Working Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to london-hackspace-metal-w...@googlegroups.com.

Des Quilty

unread,
Oct 20, 2019, 10:24:23 AM10/20/19
to London HackSpace Metal Working Group
Hi Matt,

Does our cap have ooze coming out of it? Does anybody know if we can bench test it?

Just also wondering if we could we replace the hf circuit entirely with something solide state with the help of some HS electronic experts? Would hopwdully require less maintenance and usage of expensive/unobtainable parts..

Matt Rutter

unread,
Oct 20, 2019, 10:29:57 AM10/20/19
to london-hackspace-m...@googlegroups.com
There's no ooze from it but the plastic case is cracked all the way around. 

--
https://groups.google.com/forum/?hl=en-GB#!forum/london-hackspace-metal-working-group
---
You received this message because you are subscribed to the Google Groups "London HackSpace Metal Working Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to london-hackspace-metal-w...@googlegroups.com.

Dean Forbes (PERS)

unread,
Oct 20, 2019, 10:49:54 AM10/20/19
to London HackSpace Metal Working Group
I am sure I have seen an ESR meter at the LHS lab


On Tuesday, 8 October 2019 12:31:52 UTC+1, Matt Rutter wrote:

Jonathan Hodgson

unread,
Oct 20, 2019, 3:08:42 PM10/20/19
to London HackSpace Metal Working Group
In my experience when caps fail they tend to go completely short, or completely open circuit. So a multimeter is usually enough to tell you if it's dud.


On Sunday, October 20, 2019 at 3:29:57 PM UTC+1, Matt Rutter wrote:
There's no ooze from it but the plastic case is cracked all the way around. 

Matt Rutter

unread,
Oct 20, 2019, 3:43:06 PM10/20/19
to London HackSpace Metal Working Group
The cap in question is a mica cap, 0.002uF at 5000V. it's a large thing, mounted down by the spark gap block... it's about 60x30x30.  The genuine part can be sourced, from the US for $142 plus shipping.  I found an exact equivalent, described as a Transmitting Capacitor for about £45 but that had the distinct disadvantage of being in Peru!

I did find 8off 800pF 5000V mica capacitors on ebay, in the UK for a quid each, so I bought them... 4 of them in the right combination of series and parallel will make 0.002uF... so hopefully it's a nice cheap fix.

Matt

Nigel Worsley

unread,
Oct 20, 2019, 3:46:57 PM10/20/19
to london-hackspace-m...@googlegroups.com
On Sun, 20 Oct 2019 at 20:08, Jonathan Hodgson <j.r.h...@gmail.com> wrote:
> In my experience when caps fail they tend to go completely short, or completely open circuit. So a multimeter is usually enough to tell you if it's dud.

That tends to be the eventual failed state, but a reduction in
capacitance prior to that is common. Another relevant failure mode is
breakdown at high voltage, but that is not very common for film
capacitors. The space probably has a meter that can measure
capacitance, and to check for breakdown I can do tests up to 15kV at
work.

If the capacitor is part of a tuned circuit (and I suspect that it is)
then even a small change in capacitance can be enough to stop things
working properly.

Nigle

Des Quilty

unread,
Oct 20, 2019, 5:15:12 PM10/20/19
to London HackSpace Metal Working Group
Worth a try for 4 quid. could we tune it more accurately with the addition of some extra small caps?

Nigel Worsley

unread,
Oct 20, 2019, 5:25:43 PM10/20/19
to london-hackspace-m...@googlegroups.com
On Sun, 20 Oct 2019 at 22:15, Des Quilty <des.q...@gmail.com> wrote:
> Worth a try for 4 quid

Indeed so, but don't be surprised if those caps can't handle the AC
current and let out the magic blue smoke!

> could we tune it more accurately with the addition of some extra small caps?

Probably can, I don't know how critical the tuning is for this sort of
stuff - it may be that the normal capacitor tolerance is close enough.

Nigle

Matt Rutter

unread,
Oct 20, 2019, 5:28:30 PM10/20/19
to london-hackspace-m...@googlegroups.com
If my £4 fix only lasts 2 minutes, it will at least demonstrate that the cap is the issue, and make the purchase of the real thing less risky.... $140 is a lot to gamble if we're not certain it's the cap.....

--
https://groups.google.com/forum/?hl=en-GB#!forum/london-hackspace-metal-working-group
---
You received this message because you are subscribed to the Google Groups "London HackSpace Metal Working Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to london-hackspace-metal-w...@googlegroups.com.

Des Quilty

unread,
Oct 20, 2019, 6:07:03 PM10/20/19
to London HackSpace Metal Working Group
yep, just reading that the frequency is around 2mhz, but varies quite a bit, does that sound right?

Tim Jacobs

unread,
Oct 24, 2019, 9:45:43 AM10/24/19
to london-hackspace-m...@googlegroups.com
I took a look at the manual for the machine yesterday - lots of
interesting points.

page 20 - "cleaning or dressing the spark gaps is not recommended, as
the material at the points is tungsten and impossible to file"

page 23 onwards goes into great detail about interference issues.
Interestingly it says it's essential to use solid metal conduit for the
power cable for at least 50 feet from the machine... "The importance of
correct installation cannot be overemphasized since case histories of
interference due to high frequency stabilized arc welding machines have
shown that invariably an inadequate installation was at fault." There's
a several pages of this stuff

Looking at the circuit diagram, I agree we should investigate the
capacitor, we need to take it out of the circuit to multimeter it. Also
looking at the diagram I'm not convinced of your conclusion Matt that
the remote amperage switch disables the big amperage control on the
front, looks to me like they go in series when you flip the switch.

If the capacitor is at fault, I reckon we should replace it with the
doorknob capacitors they use for tesla coils, there are loads of sellers
of those on ebay. I just found a listing for a lot of two 1000pF 5%,
7kV, 7KVA capacitors for 10EUR + 13EUR postage.

Tim


Matt Rutter

unread,
Oct 24, 2019, 10:15:26 AM10/24/19
to london-hackspace-m...@googlegroups.com
Hi Tim... You're nearly right about the current control (more right than I was!)... the external pot is in parallel with the bottom half of the internal one, so it's wiper will range from zero, up to the percentage set on the internal pot.

Those doorknob capacitors look more appropriate than the mica ones I got... but I can't see any harm in trying mine first...  I've ordered a plastic case to put them in to imitate the existing installation...

Matt 

--
https://groups.google.com/forum/?hl=en-GB#!forum/london-hackspace-metal-working-group
---
You received this message because you are subscribed to the Google Groups "London HackSpace Metal Working Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to london-hackspace-metal-w...@googlegroups.com.

Des Quilty

unread,
Oct 24, 2019, 6:19:42 PM10/24/19
to London HackSpace Metal Working Group
Will the caps arrive for the weekend?

Matt Rutter

unread,
Oct 24, 2019, 6:25:24 PM10/24/19
to london-hackspace-m...@googlegroups.com
I've got the caps, just not the box to put them in.. but hopefully that will arrive tomorrow or Saturday.... It would be nice to be testing at the weekend... It's up to parts arriving and this flu departing...

On Thu, 24 Oct 2019, 23:19 Des Quilty, <des.q...@gmail.com> wrote:
Will the caps arrive for the weekend?

--
https://groups.google.com/forum/?hl=en-GB#!forum/london-hackspace-metal-working-group
---
You received this message because you are subscribed to the Google Groups "London HackSpace Metal Working Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to london-hackspace-metal-w...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages