Tubes in broken spotwelder & other questions

[mike]

On 11/14/2016 8:54 PM, Clifford Heath wrote:

On 15/11/16 12:45, whit3rd wrote:
On Monday, November 14, 2016 at 2:44:55 PM UTC-8, et...@whidbey.com
wrote:
I found 220 mfd 450 caps at JustRadio.com. I'm not sure if these will
discharge fast enough.
For a welding purpose, 100l milliseconds is very fast (no functioning
capacitor you are likely to find would have too high ESR for this job).

I agree that the speed doesn't matter, but low ESR caps have wiring that
will survive higher currents. No sense having a whopping cap
with internally-fused connections.

I also have a bench-top spot welder, from my father who was an
orthodontist. It has a 110V input (so we needed a transformer
from 240V), and that feeds via a small Variac into a selenium
rectifier. I suspect the caps need replacing (again - they were
last done 25 years ago) and I have a bunch of 300V photoflash
caps from disposable cameras that I hope will suffice.

The welding contacts are an anvil and an upper contact connected
to the foot pedal, via an adjustable pressure-operated switch
that fires the contacter. That leaves both hands free to hold and
position the work. When you press down hard enough, the thing
fires, very cute.

What I don't know is how to determine the maximum energy I can
dump through the output transformer without saturating it.
Anyone know how I can tell (other than just keeping the
capacitance below the original value)?

Clifford Heath.

That's not a simple question.
The first order approximation is that CURRENT is what
creates the magnetic field that saturates the transformer.
If you put a scope current transformer on the primary,
load the secondary with the resistance and inductance of
your welding setup, crank up the voltage until you see
a sharp rise in the slope of the current pulse. That's
the too-much point. What happens when you add more caps
is dependent on the transformer characteristics.
Energy is linear in caps but quadratic in volts.

25 years ago, I was tasked with fixing an OEM forward
converter that would randomly self-destruct.
I built a fixture to synchronize the load transient with
the switching frequency. I could watch the primary current
head for the sky on the scope as the load transient crossed the
switching point.
Took it to the vendor site and proceeded
to blow up power supplies until they conceded that they had
a board layout problem to fix.

After the weld, the field will be sitting somewhere on the
B-H curve of the core. The next weld might be very dependent
on where you left it last time.

This is a page from my Unitek CD spot welder. Shows how they
reset the core.

http://imgur.com/ZeZerGx

I experimented with a microwave oven transformer battery tab
welder. I was hitting it with a timed pulse. Repeatability
was horrible.
When I synchronized the pulse with the line and gave it an
integral number of full cycles, the starting point on the
B-H curve was consistent and the welds got much more repeatable.

There was considerable discussion on exactly when you should terminate
the pulse, but mine was constrained to somewhere near zero current by
the triac.

 

[Clifford Heath]

On 15/11/16 17:56, mike wrote:

On 11/14/2016 8:54 PM, Clifford Heath wrote:
snip
What I don't know is how to determine the maximum energy I can
dump through the output transformer without saturating it.
Anyone know how I can tell (other than just keeping the
capacitance below the original value)?

Clifford Heath.
That's not a simple question.
The first order approximation is that CURRENT is what
creates the magnetic field that saturates the transformer.
If you put a scope current transformer on the primary,
load the secondary with the resistance and inductance of
your welding setup, crank up the voltage until you see
a sharp rise in the slope of the current pulse. That's
the too-much point. What happens when you add more caps
is dependent on the transformer characteristics.

Thanks, that all makes sense. I'd like a bit more capacity
if I can get it.

I'll have a look and see if it has a core reset current too.

Clifford Heath.

Energy is linear in caps but quadratic in volts.

25 years ago, I was tasked with fixing an OEM forward
converter that would randomly self-destruct.
I built a fixture to synchronize the load transient with
the switching frequency. I could watch the primary current
head for the sky on the scope as the load transient crossed the
switching point.
Took it to the vendor site and proceeded
to blow up power supplies until they conceded that they had
a board layout problem to fix.

After the weld, the field will be sitting somewhere on the
B-H curve of the core. The next weld might be very dependent
on where you left it last time.

This is a page from my Unitek CD spot welder. Shows how they
reset the core.

http://imgur.com/ZeZerGx

I experimented with a microwave oven transformer battery tab
welder. I was hitting it with a timed pulse. Repeatability
was horrible.
When I synchronized the pulse with the line and gave it an
integral number of full cycles, the starting point on the
B-H curve was consistent and the welds got much more repeatable.

There was considerable discussion on exactly when you should terminate
the pulse, but mine was constrained to somewhere near zero current by
the triac.

 

[Chris Jones]

On 15/11/2016 17:56, mike wrote:

On 11/14/2016 8:54 PM, Clifford Heath wrote:
On 15/11/16 12:45, whit3rd wrote:
On Monday, November 14, 2016 at 2:44:55 PM UTC-8, et...@whidbey.com
wrote:
I found 220 mfd 450 caps at JustRadio.com. I'm not sure if these will
discharge fast enough.
For a welding purpose, 100l milliseconds is very fast (no functioning
capacitor you are likely to find would have too high ESR for this job).

I agree that the speed doesn't matter, but low ESR caps have wiring that
will survive higher currents. No sense having a whopping cap
with internally-fused connections.

I also have a bench-top spot welder, from my father who was an
orthodontist. It has a 110V input (so we needed a transformer
from 240V), and that feeds via a small Variac into a selenium
rectifier. I suspect the caps need replacing (again - they were
last done 25 years ago) and I have a bunch of 300V photoflash
caps from disposable cameras that I hope will suffice.

The welding contacts are an anvil and an upper contact connected
to the foot pedal, via an adjustable pressure-operated switch
that fires the contacter. That leaves both hands free to hold and
position the work. When you press down hard enough, the thing
fires, very cute.

What I don't know is how to determine the maximum energy I can
dump through the output transformer without saturating it.
Anyone know how I can tell (other than just keeping the
capacitance below the original value)?

Clifford Heath.
That's not a simple question.
The first order approximation is that CURRENT is what
creates the magnetic field that saturates the transformer.

Be careful, remember there is a secondary current and its magnetic field
will largely cancel that of the primary, so looking at the primary
current is a red herring unless the secondary is open-circuit.

The rate of change of flux in the core will be proportional to the
primary voltage. What you need to look at is actually the integral of
primary voltage with respect to time, as this will give you the change
in the flux in the core. The longer the pulse lasts, the less voltage it
will take to cause saturation. (For similar reasons, 60Hz mains
transformers will saturate at a lower voltage if used on 50Hz mains.)
You could get more energy into the weld if you increse the voltage of
the primary pulse and decrease the duration, because the dissipation in
the weld is proportional to the integral of v sqared with respect to
time, whereas the change in the flux in the core is proportional to the
integral of v (not squared) with respect to time. Also, make sure that
the core starts out with maximum flux in the opposite direction. This
might be achieved with a reset pulse as others have pointed out,
provided the core material is one that will retain the magnetisation. If
not, you could perhaps feed a reverse current through the primary using
a low voltage power supply, to build up reversed flux just before the
main welding pulse. That might make the switching a bit more complicated.

 

On Mon, 14 Nov 2016 17:21:30 -0800, John Robertson <spam@flippers.com>
wrote:

On 201611/14/ 2:45 PM, etpm@whidbey.com wrote:
On Mon, 14 Nov 2016 12:20:23 -0800 (PST), "pfjw@aol.com"
pfjw@aol.com> wrote:

Forgive the top posting but WARNING: if a big honking electrolytic capacitor that is more than 60 years old is rattling when shaken, it is absolutely and irredeemingly, irrevocably TOAST. As are its siblings. Reforming is an exercise in futility, and possibly very dangerous, as if by some chance you do actually reform one (or more) for the moment, and it should fail under actual use, all the work previous would be for naught. **POW**.

Agree on the photoflash caps in any case. However, I suspect that the OEM caps are vastly oversize to make up for their rather slow chemistry.

Peter Wieck
Melrose Park, PA

On Monday, November 14, 2016 at 2:57:48 PM UTC-5, John Robertson wrote:
On 201611/14/ 11:12 AM, pfjw@aol.com wrote:
First, where are you? All the tubes you mention are quite common and not very expensive. I probably have all of them in my rather random connection. The bigger issue are those capacitors.

Please note the interpolations.

On Monday, November 14, 2016 at 1:34:55 PM UTC-5, et...@whidbey.com wrote:
All,
I know almost nothing about tubes(AKA valves). My son bought a
capacitor discharge spotwelder that uses tubes. It doesn't work. Parts
of it do but on the whole it appears that it doesn't charge the caps.
There are three types of tubes in the thing.
3 each OA2WA voltage regulators
2 each 2D21 thyratrons
1 each 5U4GB rectifier
There are also 6 each physically large electrolytic caps that are,
I believe, rated for 450 volts. They are made by Sprague and are
marked 200-450. No uf markings though. Apparently a couple rattle when
shaken-probably not a good sign.

200uF @ 450V. You could pay anywhere from $15 to $100 for those caps. Given the application, do not cheap out, but get as high a quality as you can afford. You also have the choice to put caps in parallel to make the capacity - and series/parallel to make a higher working voltage. There is plenty of room given the size of modern caps to go this route.


There are two transformers. The first has many taps and the second
is connected through a relay to the bank of caps. The secondary of
this xmfr is the spotwelder output.
As near as I can tell the circuit basically operates like so: The
first xmfr provides 450 volts AC which goes through the 5U4GB tube to
be rectified and then through three OA2WA tubes which are connected
in series. From there the power goes to the cap bank and charges the
caps. And from the caps through a big relay to the output xmfr.
The OA2WA tubes have only two of the pins connected. So power goes
in one pin and then out the other to the next OA2WA tube and so on. It
looks like there is no output from the last tube in the series. So
maybe this is why the caps won't charge.


I suggest you consider PHOTOFLASH rated caps - they have a very fast
discharge (low internal ESR - you might want to get a Blue ESR meter to
test these caps) and aren't too expensive...

http://ca.mouser.com/Passive-Components/Capacitors/_/N-5g7r?Keyword=photoflash&FS=True

Look up REFORMING capacitors. Even old ones like yours may be saved,
there was a recent thesis published that showed how large value caps can
last a long time if they are reformed carefully.

2D21's are very old design - there are many circuits online to replace
them with SCRs if you want to bring the machine up to the 21st century...

Likewise your 5U4s, this can be replaced with a solid state bridge (dual
diode) rectifier. As the 5U4 has its own transformer then you need to
make sure it is fused correctly on the primary side in case of a shorted
diode.

OA2 tubes are 150VDC regulators, so with three of them the output is
regulated to 450VDC. If the OA2s are glowing purple then they are at
their regulated voltage (more or less) - all three need to be glowing
for your 450VDC to be regulated. I would expect them to be across the
output of the 5U4 to ground to regulate the output voltage...

John :-#)#


It is probably a blessing that your rectifier (5U4) tube does not work. Were you to be passing B+ - well, the results may be unhappy - see below.

At least one 2D21 thyratron seems to work because the relay that
connects the cap bank works when the machine is triggered.
I would like to know how to test the tubes without a tube tester.

Other than for filament continuity and gross shorts (shorts on a cold tube), no you have no certain way to test tubes without a tester.

Maybe this means I will need to make a tester. I also would like to
know how to test the big caps. My meter has a cap test function but it
doesn't seem to work with these big caps. Maybe this is because the
caps are high voltage types. Is it (fairly) safe to attempt charging
the caps with line voltage (120 volts) by first rectifying and then
passing the current through a resistor and then to the caps?
Here is a link to pics my son took of the guts of the spotwelder:
https://goo.gl/photos/LF4JKeGyA4YWiCwv9
Any comments about the spotwelder would be much appreciated.
Thanks,
Eric

The big caps are toast - full stop. DO NOT EVEN TRY to operate the tester without them as it is possible to do permanent damage to the transformers if the caps are shorted.

If you are anywhere near the Melrose Park section of Cheltenham Township (Philadelphia area) let me know. Despite the size and industrial application, this is a fairly basic issue to troubleshoot.

Now, showing my age, I used one of those ancient beasts some 40 years ago when I was working as a machinist. When set up properly, they are amazingly versatile. I used one to weld various thickness of inconel wave washers and pimple washers onto stainless steel seal backs for aircraft and nuclear applications. From stuff that was almost as thin as paper (0.004 thick) to stuff as thick as index card stock - dial in the machine, pass the pull-tests and it would hold the setting, effectively, until changed. There was a small stash of tubes above the machine - but no 2D21s. When one of them died one way, I found it at Leon Fertik's establishment for $1.50. I replaced them both and purchased two spares. The machine shop is still with us and thriving. Leon, sadly gave up his shop after his wife passed.

Best of luck with it - I can think of a lot of uses.... Especially if your son is an artist.

Peter Wieck
Melrose Park, PA



--
(Please post followups or tech inquiries to the USENET newsgroup)
John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9
(604)872-5757 or Fax 872-2010 (Pinballs, Jukes, Video Games)
www.flippers.com
"Old pinballers never die, they just flip out."
I found 220 mfd 450 caps at JustRadio.com. I'm not sure if these will
discharge fast enough. I emailed the site and am waiting for an
answer. A quick search for photoflash caps turns up 300 volt caps, no
450 volt caps. So I'll wait to see if Just Radio thinks its caps are
up to the job. I have decided to just change out all the caps in the
thing if the correct main big caps are available at a reasonable
price. I don't think I want to go through the trouble of coming up
with solid state solutions and then rewiring the unit just yet. If it
worked that well in the past with tubes and if they have reasonable
lifetimes I'll just keep it the way it is. On the other hand, it is my
son's machine and he can do whatever he wants with it, no matter what
kind of help dad wants to give.
Eric


Did you look at the Mouser link I included a few posts ago? They list
400 and 500VDC Photoflash caps. Mind you, you need to order one to two
hundred.

http://ca.mouser.com/Passive-Components/Capacitors/_/N-5g7r?Keyword=photoflash&FS=True

Here is a surplus place with photoflash caps, not cheap, but they are in
stock:

https://www.surplussales.com/capacitors/Electrolytics/eaf.html

However, if you use the part number you will probably find a stocking
distributor...

JustRadio sells good caps, but they aren't photoflash rated - pretty
safe bet on that!

John :-#)#

I can't use 100 caps. I did find photoflash caps but the leads are
kinda thin. I'm still waiting to see if Just Radio thinks their caps
will do.
Eric

 

On Mon, 14 Nov 2016 17:45:28 -0800 (PST), whit3rd <whit3rd@gmail.com>
wrote:

On Monday, November 14, 2016 at 2:44:55 PM UTC-8, et...@whidbey.com wrote:

I found 220 mfd 450 caps at JustRadio.com. I'm not sure if these will
discharge fast enough.

For a welding purpose, 100l milliseconds is very fast (no functioning
capacitor you are likely to find would have too high ESR for this job).

ESR gets to be important with low voltage high current, where a fraction of
a volt of ripple is a malfunction. Fraction of a volt difference on a 450V capacitor
charge is negligible.

So photoflash rating is probably not necessary?
Thanks,
Eric

 

On Mon, 14 Nov 2016 22:56:14 -0800, mike <ham789@netzero.net> wrote:

On 11/14/2016 8:54 PM, Clifford Heath wrote:
On 15/11/16 12:45, whit3rd wrote:
On Monday, November 14, 2016 at 2:44:55 PM UTC-8, et...@whidbey.com
wrote:
I found 220 mfd 450 caps at JustRadio.com. I'm not sure if these will
discharge fast enough.
For a welding purpose, 100l milliseconds is very fast (no functioning
capacitor you are likely to find would have too high ESR for this job).

I agree that the speed doesn't matter, but low ESR caps have wiring that
will survive higher currents. No sense having a whopping cap
with internally-fused connections.

I also have a bench-top spot welder, from my father who was an
orthodontist. It has a 110V input (so we needed a transformer
from 240V), and that feeds via a small Variac into a selenium
rectifier. I suspect the caps need replacing (again - they were
last done 25 years ago) and I have a bunch of 300V photoflash
caps from disposable cameras that I hope will suffice.

The welding contacts are an anvil and an upper contact connected
to the foot pedal, via an adjustable pressure-operated switch
that fires the contacter. That leaves both hands free to hold and
position the work. When you press down hard enough, the thing
fires, very cute.

What I don't know is how to determine the maximum energy I can
dump through the output transformer without saturating it.
Anyone know how I can tell (other than just keeping the
capacitance below the original value)?

Clifford Heath.
That's not a simple question.
The first order approximation is that CURRENT is what
creates the magnetic field that saturates the transformer.
If you put a scope current transformer on the primary,
load the secondary with the resistance and inductance of
your welding setup, crank up the voltage until you see
a sharp rise in the slope of the current pulse. That's
the too-much point. What happens when you add more caps
is dependent on the transformer characteristics.
Energy is linear in caps but quadratic in volts.

25 years ago, I was tasked with fixing an OEM forward
converter that would randomly self-destruct.
I built a fixture to synchronize the load transient with
the switching frequency. I could watch the primary current
head for the sky on the scope as the load transient crossed the
switching point.
Took it to the vendor site and proceeded
to blow up power supplies until they conceded that they had
a board layout problem to fix.

After the weld, the field will be sitting somewhere on the
B-H curve of the core. The next weld might be very dependent
on where you left it last time.

This is a page from my Unitek CD spot welder. Shows how they
reset the core.

http://imgur.com/ZeZerGx

I experimented with a microwave oven transformer battery tab
welder. I was hitting it with a timed pulse. Repeatability
was horrible.
When I synchronized the pulse with the line and gave it an
integral number of full cycles, the starting point on the
B-H curve was consistent and the welds got much more repeatable.

There was considerable discussion on exactly when you should terminate
the pulse, but mine was constrained to somewhere near zero current by
the triac.

Greetings Mike,
The machine I am working with is a Unitek model 1-048-03. Is this the
same as yours? If not do you know if it is close enough to use for
trouble shooting and operating the machine I am trying to get working?
If it would be useful is there some way you could get a copy to me?
Maybe scan and email? I don't mind paying you for your trouble. I
wouldn't even ask but so far I have not been able to find a manual of
any type for the model my son has.
Thanks,
Eric

 

[mike]

On 11/15/2016 11:38 AM, etpm@whidbey.com wrote:

On Mon, 14 Nov 2016 22:56:14 -0800, mike <ham789@netzero.net> wrote:

On 11/14/2016 8:54 PM, Clifford Heath wrote:
On 15/11/16 12:45, whit3rd wrote:
On Monday, November 14, 2016 at 2:44:55 PM UTC-8, et...@whidbey.com
wrote:
I found 220 mfd 450 caps at JustRadio.com. I'm not sure if these will
discharge fast enough.
For a welding purpose, 100l milliseconds is very fast (no functioning
capacitor you are likely to find would have too high ESR for this job).

I agree that the speed doesn't matter, but low ESR caps have wiring that
will survive higher currents. No sense having a whopping cap
with internally-fused connections.

I also have a bench-top spot welder, from my father who was an
orthodontist. It has a 110V input (so we needed a transformer
from 240V), and that feeds via a small Variac into a selenium
rectifier. I suspect the caps need replacing (again - they were
last done 25 years ago) and I have a bunch of 300V photoflash
caps from disposable cameras that I hope will suffice.

The welding contacts are an anvil and an upper contact connected
to the foot pedal, via an adjustable pressure-operated switch
that fires the contacter. That leaves both hands free to hold and
position the work. When you press down hard enough, the thing
fires, very cute.

What I don't know is how to determine the maximum energy I can
dump through the output transformer without saturating it.
Anyone know how I can tell (other than just keeping the
capacitance below the original value)?

Clifford Heath.
That's not a simple question.
The first order approximation is that CURRENT is what
creates the magnetic field that saturates the transformer.
If you put a scope current transformer on the primary,
load the secondary with the resistance and inductance of
your welding setup, crank up the voltage until you see
a sharp rise in the slope of the current pulse. That's
the too-much point. What happens when you add more caps
is dependent on the transformer characteristics.
Energy is linear in caps but quadratic in volts.

25 years ago, I was tasked with fixing an OEM forward
converter that would randomly self-destruct.
I built a fixture to synchronize the load transient with
the switching frequency. I could watch the primary current
head for the sky on the scope as the load transient crossed the
switching point.
Took it to the vendor site and proceeded
to blow up power supplies until they conceded that they had
a board layout problem to fix.

After the weld, the field will be sitting somewhere on the
B-H curve of the core. The next weld might be very dependent
on where you left it last time.

This is a page from my Unitek CD spot welder. Shows how they
reset the core.

http://imgur.com/ZeZerGx

I experimented with a microwave oven transformer battery tab
welder. I was hitting it with a timed pulse. Repeatability
was horrible.
When I synchronized the pulse with the line and gave it an
integral number of full cycles, the starting point on the
B-H curve was consistent and the welds got much more repeatable.

There was considerable discussion on exactly when you should terminate
the pulse, but mine was constrained to somewhere near zero current by
the triac.
Greetings Mike,
The machine I am working with is a Unitek model 1-048-03. Is this the
same as yours? If not do you know if it is close enough to use for
trouble shooting and operating the machine I am trying to get working?
If it would be useful is there some way you could get a copy to me?
Maybe scan and email? I don't mind paying you for your trouble. I
wouldn't even ask but so far I have not been able to find a manual of
any type for the model my son has.
Thanks,
Eric

It's a unitek 125.
All I have is a very poor schematic and three pages of
part of the cal procedure.

http://i.imgur.com/vSg8agC.png
http://i.imgur.com/mrQIq1f.png
http://i.imgur.com/lE2DroY.png
http://i.imgur.com/8Nv2KpU.png
http://i.imgur.com/TZyfuLp.png

If you have a way to archive them permanently, please do so.
They won't stay on my page for long.

This is my DIY battery tab welding fixture made out of an arbor press.

http://i.imgur.com/er1BqSb.jpg
http://i.imgur.com/OeQZdWH.jpg

I emailed Unitek years ago. At that time, they were very helpful.

 

[pfjw@aol.com]

OK!! EVERYBODY!!

Step back a minute and have a look at what is being discussed here:

a) This is a TUBE-based spot welder designed probably more than 60 years ago.
b) The bean counters (a scourge even then) would NEVER have specified photo-flash capacitors when just a bit more of conventional capacitance would address 'slow chemistry' at a fraction of the cost.
c) A healthy gap on the transformer would eliminate sateuration issues in this application - 'at no additional cost' and without complications, and entirely permanently.
d) This is a piece of industrial equipment designed to be used by production worker who very likely have neither the time nor the inclination to do anything other than follow directions, set the various knobs and switches then GO for hours at a time.
e) And, as with any decent tool from any decent manufacturer, it is designed to be serviced in a simple and straightforward manner by techs who need do no more than follow directions from a standard troubleshooting schedule.
f) And, finally on this list, replacement parts, especially generics (capacitors, tubes) will be entirely conventional - no silliness about "matched pairs" for instance, or boutique caps. If one has a small machine shop in Bucksnort, Tennessee ( https://en.wikipedia.org/wiki/Bucksnort,_Tennessee ) one does not have the time to search down something exotic. One is going to get on the phone to Graybar and order whatever it says on the can, and Graybar will have it in stock. http://www.graybar.com/ And, back in the day Graybar stocked industrial tubes as well.

Meaning that there is NOTHING esoteric about this device, and it will not require ANY special 'stuff' to make it go. The caps in this device are toast. If even one of them "rattles" then all of them, being of similar age, living in similar environments and doing similar work, are also done. Not replacing all of them is just plain stupid. Full Stop.

Purchase physically rugged caps that either are, or add up to, 200uF & 450 VDC per each, at least. Make sure that the soldering points and connections can handle anticipated voltages and currents. Replace all the tubes, even if they are not necessarily done, as now you will be starting fresh, nor is the investment all that much. If I have 2D21s, I would send the whole shebang to Oregon/Washington State for US$30 including shipping and think I was making too much. Otherwise, AES will do the same for perhaps half-again as much.

Sheesh, guys and gals - it is almost as if this were rocket science. It ain't nohow.

Peter Wieck
Melrose Park, PA

 

[Geo]

On Mon, 14 Nov 2016 10:38:12 -0800, etpm@whidbey.com wrote:

<snip>

The OA2WA tubes have only two of the pins connected. So power goes
in one pin and then out the other to the next OA2WA tube and so on. It
looks like there is no output from the last tube in the series. So
maybe this is why the caps won't charge.
snip

I cannot see properly from the photos - but is there a connection to
the last (bottom) regulator cathode?
I would guess it might be to chassis so "no output" is correct.

If all three glow then the voltage across each should be fine.

 

[pfjw@aol.com]

On Wednesday, November 16, 2016 at 12:06:15 PM UTC-5, et...@whidbey.com wrote:

Please note the interpolations:

That's it, I'm convinced. I'm gonna order the caps from JustRadio.
They are the correct rating for voltage and capacitance, different
connections-not screws, but big tabs instead. Easily soldered to the
bus bars. I have been thinking about those big bus bars and am
wondering if the size is either for mechanical reasons or lower
resistance reasons.

Both. Mechanical as this machine moves, vibrates and is subject to various stresses thereby. And, consider that the shop I worked in a _long_ time ago used their welder two 10-hour shifts in a row, about 2 days per week. That is a lot of welds.

Electrical as the enemy of these machines is heat. Heat increases resistance which can throw off settings - and with 0.003 inconel pimple washers the difference between the weld not taking and a burn-through isn't much, and on aircraft and/or nuclear rated parts that is unacceptable.

I can't see the original caps dumping enough

current to need such robust bars. Especially considering the relay
contacts connected to the cap bank and the much lighter wires running
from the bus bars and relay to the xmfr. Lemme know if you have all
the tubes and I'll send a check.

You will have to wait into the weekend for me to get an edgewise to check my stock. Our youngest cat (3 years old, 18 pound Maine Coon mix) has just discovered that he can use the dog door. So, my primary task is to cat-proof the dog yard so he may go outside, but not any further. We may be suburban, but we have foxes, raccoons, other cats and cars to contend with.

As to sending a check, that is not how it works. If I have the tubes, I will send them to you - and then *if* they work out, you may send the check.

Peter Wieck
Melrose p

 

[pfjw@aol.com]

On Wednesday, November 16, 2016 at 1:04:24 PM UTC-5, John Robertson wrote:

Keep your old tubes until you can test them and be sure they are bad.
These old tubes are not getting any more common! 2D21s were always a bit
fussy and were replaced by 2050 tubes not long after they came
out...even new out of the box there were failures. Seeburg jukeboxes
used 2D21s for their early tormat design and quickly changed the design
to 2050s part way through the second year of production due to read-out
and write-in issues. I have schematics for 2D21 testers that I can dig
up (probably find similar schematics easily enough online) and post of
my tech ftp pages if there is a demand/request. The 2D21 is pretty easy
to test, if you have access to a good tube tester it will usually show
them on the Seldom Used auxiliary tube list. The 2D21s are go/no-go
kinda things.

OA2s are also pretty basic tubes. They use a conductive gas to regulate
to 150VDC. If the tube is not broken or gassy then it should be just
fine. Check the Getter flash in the tube - it should not be showing any
white powder, that is a tube that has lost its seal and oxygen has
entered...

2D21s are/were used commonly in dental X-ray and Fluoroscope machines, and when, also back in the day, I was working for a GC doing some serious rebuilding at HUP in Philadelphia, one of the docs needed to resurrect such a machine for historical purposes. I found him the tubes he needed locally, including two 2D21s. This was "Before Internet", so not an instantaneous process.

I also keep a very good tester (Hickok 539B), so no tube would leave this house untested.

Yes, of course, keep any undamaged old tube.

Peter Wieck
Melrose Park, PA

 

On Wed, 16 Nov 2016 09:25:37 +0000, Geo <nhhu-o3hu@dea.spamcon.org>
wrote:

On Mon, 14 Nov 2016 10:38:12 -0800, etpm@whidbey.com wrote:

snip
The OA2WA tubes have only two of the pins connected. So power goes
in one pin and then out the other to the next OA2WA tube and so on. It
looks like there is no output from the last tube in the series. So
maybe this is why the caps won't charge.
snip

I cannot see properly from the photos - but is there a connection to
the last (bottom) regulator cathode?
I would guess it might be to chassis so "no output" is correct.

If all three glow then the voltage across each should be fine.

There is a connection. It does not go to the chassis. I think the tube
is bad. All tubes are gonna be replaced anyway.
Eric

 

On Wed, 16 Nov 2016 05:35:40 -0800 (PST), "pfjw@aol.com"
<pfjw@aol.com> wrote:

OK!! EVERYBODY!!

Step back a minute and have a look at what is being discussed here:

a) This is a TUBE-based spot welder designed probably more than 60 years ago.
b) The bean counters (a scourge even then) would NEVER have specified photo-flash capacitors when just a bit more of conventional capacitance would address 'slow chemistry' at a fraction of the cost.
c) A healthy gap on the transformer would eliminate sateuration issues in this application - 'at no additional cost' and without complications, and entirely permanently.
d) This is a piece of industrial equipment designed to be used by production worker who very likely have neither the time nor the inclination to do anything other than follow directions, set the various knobs and switches then GO for hours at a time.
e) And, as with any decent tool from any decent manufacturer, it is designed to be serviced in a simple and straightforward manner by techs who need do no more than follow directions from a standard troubleshooting schedule.
f) And, finally on this list, replacement parts, especially generics (capacitors, tubes) will be entirely conventional - no silliness about "matched pairs" for instance, or boutique caps. If one has a small machine shop in Bucksnort, Tennessee ( https://en.wikipedia.org/wiki/Bucksnort,_Tennessee ) one does not have the time to search down something exotic. One is going to get on the phone to Graybar and order whatever it says on the can, and Graybar will have it in stock. http://www.graybar.com/ And, back in the day Graybar stocked industrial tubes as well.

Meaning that there is NOTHING esoteric about this device, and it will not require ANY special 'stuff' to make it go. The caps in this device are toast. If even one of them "rattles" then all of them, being of similar age, living in similar environments and doing similar work, are also done. Not replacing all of them is just plain stupid. Full Stop.

Purchase physically rugged caps that either are, or add up to, 200uF & 450 VDC per each, at least. Make sure that the soldering points and connections can handle anticipated voltages and currents. Replace all the tubes, even if they are not necessarily done, as now you will be starting fresh, nor is the investment all that much. If I have 2D21s, I would send the whole shebang to Oregon/Washington State for US$30 including shipping and think I was making too much. Otherwise, AES will do the same for perhaps half-again as much.

Sheesh, guys and gals - it is almost as if this were rocket science. It ain't nohow.

Peter Wieck
Melrose Park, PA

That's it, I'm convinced. I'm gonna order the caps from JustRadio.
They are the correct rating for voltage and capacitance, different
connections-not screws, but big tabs instead. Easily soldered to the
bus bars. I have been thinking about those big bus bars and am
wondering if the size is either for mechanical reasons or lower
resistance reasons. I can't see the original caps dumping enough
current to need such robust bars. Especially considering the relay
contacts connected to the cap bank and the much lighter wires running
from the bus bars and relay to the xmfr. Lemme know if you have all
the tubes and I'll send a check.
Thanks,
Eric

 

[John Robertson]

On 2016/11/16 9:04 AM, etpm@whidbey.com wrote:

On Wed, 16 Nov 2016 09:25:37 +0000, Geo <nhhu-o3hu@dea.spamcon.org
wrote:

On Mon, 14 Nov 2016 10:38:12 -0800, etpm@whidbey.com wrote:

snip
The OA2WA tubes have only two of the pins connected. So power goes
in one pin and then out the other to the next OA2WA tube and so on. It
looks like there is no output from the last tube in the series. So
maybe this is why the caps won't charge.
snip

I cannot see properly from the photos - but is there a connection to
the last (bottom) regulator cathode?
I would guess it might be to chassis so "no output" is correct.

If all three glow then the voltage across each should be fine.
There is a connection. It does not go to the chassis. I think the tube
is bad. All tubes are gonna be replaced anyway.
Eric

Keep your old tubes until you can test them and be sure they are bad.
These old tubes are not getting any more common! 2D21s were always a bit
fussy and were replaced by 2050 tubes not long after they came
out...even new out of the box there were failures. Seeburg jukeboxes
used 2D21s for their early tormat design and quickly changed the design
to 2050s part way through the second year of production due to read-out
and write-in issues. I have schematics for 2D21 testers that I can dig
up (probably find similar schematics easily enough online) and post of
my tech ftp pages if there is a demand/request. The 2D21 is pretty easy
to test, if you have access to a good tube tester it will usually show
them on the Seldom Used auxiliary tube list. The 2D21s are go/no-go
kinda things.

OA2s are also pretty basic tubes. They use a conductive gas to regulate
to 150VDC. If the tube is not broken or gassy then it should be just
fine. Check the Getter flash in the tube - it should not be showing any
white powder, that is a tube that has lost its seal and oxygen has
entered...

John :-#)#

--
(Please post followups or tech inquiries to the USENET newsgroup)
John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9
(604)872-5757 or Fax 872-2010 (Pinballs, Jukes, Video Games)
www.flippers.com
"Old pinballers never die, they just flip out."

 

[Clifford Heath]

On 17/11/16 00:35, pfjw@aol.com wrote:

OK!! EVERYBODY!!
Step back a minute and have a look at what is being discussed here:
Sheesh, guys and gals - it is almost as if this were rocket science.

You're responding to Chris' answer to *my* question about how I
can make *my* welder do more than it was designed to. I'm thankful
to him for that, and I don't think he needs to "step back a minute".

 

[pfjw@aol.com]

On Wednesday, November 16, 2016 at 4:30:36 PM UTC-5, Clifford Heath wrote:

On 17/11/16 00:35, pfjw@aol.com wrote:
OK!! EVERYBODY!!
Step back a minute and have a look at what is being discussed here:
Sheesh, guys and gals - it is almost as if this were rocket science.

You're responding to Chris' answer to *my* question about how I
can make *my* welder do more than it was designed to. I'm thankful
to him for that, and I don't think he needs to "step back a minute".

No, I made a generic suggestion that we not over-analyze what is, in actuality, a very simple issue. A discussion about cap-reforming has no place in this particular discussion, nor do photo-flash caps except as a perhaps interesting sidebar. A "you might try this if practical" type thing. And at that level, I agree, one might try it.

This discussion, for the OP in any case, is about restoring a piece of industrial equipment back to reliable and optimal function. Emphatically Not rocket science.

Now, here is the issue with industrial tools with the goal of making it 'do more than it was designed to do'. Well designed tools have a level of resilience already built in. Going beyond those parameters puts the tool, operator and surroundings at-risk either a little bit, or perhaps quite a bit depending on the tool. It is, in every case, under every condition, an exceedingly bad idea. If you choose to put yourself and your property at-risk, that is your privilege - and may you never be in line for a Darwin Award.

I have worked in an industrial production environment with very fast, very powerful, very hot machines, materials and moving parts in close proximity. I would NOT want to think that some lashed-up tool operating outside its design parameters had my life or health dependent upon its smooth operation.

Peter Wieck
Melrose Park, PA

 

[pfjw@aol.com]

On Friday, November 18, 2016 at 7:02:02 AM UTC-5, Chris Jones wrote:

meh. That is quite a generalisation. Sometimes upgrades and
modifications make things safer, especially when starting with older
equipment that, in its original condition, would not meet modern safety
standards. Of course one has to apply sensible engineering, and not
overstress parts, and consider any safety consequences of parts failing,
but that does not necessarily preclude improving performance in some
cases, as the original designers of the equipment did not have the
components available to them that we do now.

Sure. But, look at what you just stated. Sensible engineering. Not willy-nilly speculation by individuals with no or very limited basic knowledge of the design parameters of the tool/machine involved.

"more than they were designed to do" is not sensible engineering.

Better bearings, better capacitors, better lubricants, more effective switches, better snubbers (rubber parts), better insulation, better electrodes, better shielding - all of the above are perfectly valid (and sensible) improvements for about anything. One does not have to use 30w non-detergent oil in a vintage automobile, or propylene glycol antifreeze today.

But, at the same time, running a stock Model T on nitro-methane is, perhaps, inadvisable.

Peter Wieck
Melrose Park, PA

 

[Chris Jones]

On 18/11/2016 02:31, pfjw@aol.com wrote:

Well designed tools have a level of resilience already built in.
Going beyond those parameters puts the tool, operator and
surroundings at-risk either a little bit, or perhaps quite a bit
depending on the tool. It is, in every case, under every condition,
an exceedingly bad idea.

meh. That is quite a generalisation. Sometimes upgrades and
modifications make things safer, especially when starting with older
equipment that, in its original condition, would not meet modern safety
standards. Of course one has to apply sensible engineering, and not
overstress parts, and consider any safety consequences of parts failing,
but that does not necessarily preclude improving performance in some
cases, as the original designers of the equipment did not have the
components available to them that we do now.

 

[Ron D.]

I totally agree. I rebuilt a nearly 40 YO gas dryer:

I upgraded the drum lighting to LED. I upgraded the kind of thrust bearings. I added two thrust bearings. I bought a new old stock bearing assembly for the fan and upgraded the grease. I painted and replaced the new nylon hinges on the filter door. Nylon and heat don't like each other. I painted the rusty side that was next to the washer. I replaced the 40 YO drum belt. I replaced the blower belt. I upgraded/replaced, made from scratch a duct gasket. The thrust bearings were $24.00 not including postage.

I changed a set screw to brass tipped so the pulley would slip if the bearing housing froze. Now a cool upgrade would be a belt slip detector.

All of these are minor upgrades, but they make it better.

I have a few more planned:

1) Make a new lint filter
2) Make it easier to do preventative maintenance by adding Threadserts instead of sheet metal screws.

The upgrades were costly. The grease was about $30.00. The material to make the filter was about $70.00, but I'll use that also to make the lint filter. The threadsert kit was about $80.00 USD.

I will try to measure and determine the type of wire in the ignitor and maybe eventually make a replacement. I have two brand new ones.