[Prev][Next][Index][Thread]
Re: Various Transformers-questions
Hi Andy,
Comments interspersed and below.
> Original Poster: "Andy Cleary" <gemware2-at-dreamscape-dot-com>
> I unpotted an old 15kv 60ma transformer using the kerosene soak
>method. It had been soaking for a few months (I previously didn't
>have time to do anything with it), so I figured I would try to finish
> There was certainly a break in one side of the HV, and I unwound
>a few layers on each to provide matching resistances. In the
>process I also removed 4 metal shunts from each stack. Was
>that too many?
That is a hard question to answer based on the info you give,
here. How many shunts are there total? How massive is the
xformer core? You know that I have increased the current of
my 75mA xformer by a *subtle* ;o) 467% w/o losing the NST.
>I have since tried the transformer, and it functions well. Using
>a spark-gap voltage test, and even with a variac test, my
>results imply that it is 17000 volts, which cannot be. How
>would I find the new voltage (and current) specifications?
Hmm. This test won´t tell you anything, other than that the
xformer produces HV. If you remember my post to Scot and
Jon, you can´t measure the voltage this way. In order to
determine the difference between 15kV & 17kV with a spark
gap, you would need a very controlled environment. You need
to know the actual breakdown voltage of the gap, which
depends on the humidity, the atmospheric pressure, just to
name a few (unknown) variables. Best thing is to find an AC
high voltage probe and measure it, but BE CAREFUL. It is
a dangerous task and in my opinion you don´t really need to
know it.
> Could I connect a ma meter between the leads (thus
> shorting them and envoking the current limiting?).
As long as your xformer still has shunts OR it limits the
current through core design, you can use a DMM set to
the 20A AC scale. Use ONLY the unfused 20A scale to
measure. Connect the DMM across the secondary. It
would be best to elevate the DMM on something that
insulates. Don´t touch the meter, while the xformer is
running. Next connect a second DMM across the primary
windings (voltage) and SLOWLY ramp up your variac, while
reading (best to get a helper that does this) the primary
voltage and watching for signs of any failure. As soon as
you have reached the operating voltage, read the mA off
your first DMM.
Do the acid tests:
1.) Remove all the meters. Short out your xformer´s
secondary with a wire. Connect one DMM to measure
the primary voltage and the second one (20A) scale in
series with your variac and xformer primary. Ramp up
the variac and watch the meters. If everything seems
sound, let her rip for 3-5 minutes. Shut down and unplug
everything and finger probe the xformer. If it is hot, forget
it, it will burn out sooner or later. If not, reconnect it and
run it for 30-60 minutes, while taking a look at the meters
once in a while. If the xformer stays cool to the touch or
only slightly warm, you have succeeded. On to the next
step:
2.) Remove all meters and the secondary short. Use some
heavy wire or 1/16" copper tubing and bend yourself a
Jacob´s Ladder. Run the unit for 5-15 minutes. If it survives
this, too, you can say you have a HV xformer that is just
begging to be used in a TC ;o))).
> Also, when I put the three coils back on the core, should
>all the windings be wound in the same direction?
They should be put back on, the way the were before you
stripped them down.
>Can I parallel another "fixed" transformer in the future?
Yes, sure, why not?
>Would such a transformer supply be suitable with an 8"
>coil? Could I connect it in parallel with a 12/60 or
> 15/30?
I would max out this one first. If you parallel, then I would
go for the 15/30 first. If you add the 12/60 later (for a total
of 3 xformers) go back to the steps above and measure it
all again. That way you KNOW what current your setup will
produce. The 15/30 and the 12/60 alone will get you 1100VA,
which should be good for some 50+" sparks. If your modified
15/60 gets you 1500VA (which I think is possible), you will
have a total of 2.6kVA, which would increase your spark
length to about 80-85", if you take the time to tweak it.
> As another alternative, I have been working on a MOT
>supply, as featured on the altair-dot-org web site. MOTS are
>much easier to come by than NSTs, and seem to provide
>a good current output.
MOTS are DEADLY. The schematic you mentioned was
worked out by William Payne and me. I posted my thoughts
and comments on the list and he drew up this schematic as
a result of those posts. You might want to search the
archives on it. I don´t remember, when I posted them. It
must have been sometime early 1998, I think. Short comments
below (been there, done that=> lots of problems):
- MOTS are not designed to be seriesed. If you build the
"isolation xformers and PSU" out of MOTS, you need a LOT
of EXACTLY the SAME MOTS (not just similar values, the
SAME type, manufacturer and it would be best if they were
from the same lot) > almost impossible, because not even
the µwave manufacturer can tell you what MOTS are used
in µwave type "x". They use the cheapest that they can get
at the time of µwave manufacture.
- MOTS are NOT designed to run in a short circuit config. The
magnetron doesn´t ever present a total short, which is why
the manufacturer can skimp on shunt design. MOTS MUST
be externally limited.
- I did some extreme testing and found only ONE type suitable
for coiling usage (out of 15 different types tested). They were
2.3kV-at-650mA from Gardner in the U.K. (which is good for
you, as you live in the U.K.). I could series 4 of these, out of
oil, w/o any corona or flashover, so I was planning on using
two pairs of 3 in series under oil. I would have grounded the
"center tap" between the two pairs. This would have given me
13.8kV-at-650mA, which IS pole pig power (~9kVA). However,
I was only able to find a total of 4 of these.
- In order to adequately protect a seriesed MOT PSU, you MUST
use a safety gap across EACH AND EVERY MOT. A sg across
the total number of MOTS does NOT protect all MOTS. They
WILL die, if a kickback occurs. I learned that the hard way ;o)
Adjusting such a multiple series safety gap (MSSG) is a PIA
to say the least.
- As one of the wires is connected to the core (don´t even think
about removing this connection. It DOESN´T work), you will
need to design your setup, so that the xformers aren´t in danger
of flashing over to the core and back to the primary (a VERY
dangerous situation) or to any grounded object.
In my opinion 2 seriesed, center tapped and grounded, MOTS
are great for a toob TC (TTC), but nothing for the pulse powered
coils, we build.
> All of the transformers are the same in the parallel set, do I
>have to worry about connecting the MOTS on the right phase
>(if that is the term)?
Yes, all xformers must be phased correctly, so that the outputs
add up. This might have been your problem.
>Would a DC type MOT supply be better (the voltage doubler
>with capacitors and diodes)?
Well, if you double the voltage (it is more of a level shifter, than a
doubler), you double the voltage and half the current, so the VA
is the same. Nothing to be gained really AND you need caps that
are big enough, so that the (doubled) voltage doesn´t collapse
during recharge. It is more of a trade-off than an advantage to
my mind.
>How would a DC supply effect coil design?
More complicated. You would need to protect the diodes from
too high a PIV. The only advantage, that I can see in a DC-TC
is the fact that you can use a variable speed async rotary to
control your feeding power.
In my opinion stay away from MOTS. Start with your NST setup
and go up in power as you get more confident with HV. They
WILL provide you with enough fun for starters. Why not jump on
the U.K. pole pig buy. Those 5kVA suckers can be run at 10,
maybe even 15kVA, for short runs. More than enough power
(for me at least) ;o)). However DO visit:
http://www.drmegavolt-dot-com/underpages/warning.asp
FIRST and be sure, a MOT PSU can BBQ you in a similar fashion.
Hope this helped you.
Coiler greets from Germany,
Reinhard