[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] The Dreadful Task of Ballasting (longish)

Hi Phil,
My gut feeling is that resonance is not the problem. It may be an inductive kick, but possibly it is simply due to cap charge and time. Are you running a rotary gap?
I'll forward you an Excel spreadsheet I use now and then for looking at various aspects of a transformer. I typically run a series of measurements at various voltages logging open and short circuit currents, voltages, etc., but this can be done at a single voltage of course. I think you may be better off attempting to attack some of this data so that your sure about the inductances and about the true Cres of the transformer. If you give the spreadsheet good data, it will tell you inductances, reactances, leakage inductances, leakage reactances, mutual inductance, coupling, and actual Cres, based on measurements. After a new build like this, these are things I would want to know for sure about the transformer.
But regarding your problem. The cap appears to already be very STR even at 43.2nF. I did some back calcs on your given inductances and it looks like Cres is around 120nF. But this also assumes all the data is good. Added ballast inductance will be reflected to the secondary side and will of course alter Cres, but I think you are still very STR, so if it's a rotary and possibly the time to alignment was slow, the safety gap may simply be reaching breakdown due to cap charging. The new transformer certainly gives you 3X the charging current, so it's reaching potential that much faster and if there nothing to break to, then it will continue to charge beyond the 11kV guestimate until the safety gap breakdown voltage is reached. 

Take care,
Bart


Regarding the

Phil Tuck wrote:

Hello.

The last two days have been spent trying to successfully ballast my new
tranny, yet I seem to come up against the same obstacle. Resonance with the
ballast I think


Specifications:

240v 50Hz Tranny giving 11K 350m/a  (the 11K is an estimate based on low
voltage projections) 350m/a is measured.

Cp  = 43.2nF (measured) or if I remove one string 32.4nF.

I have two  temporary ballasts, an MOT at 53.4mH and a welder at 60.5mH

The 10amp Variac is 0.58H - so far as you can measure these accurately with
a DVM

Tranny primary is 176 mH (sec open) & 10.4 mH (sec shorted)
Turns ratio squared is 1764 
I decided on a 16 amp primary draw @ 270v on the Variac, so 270/16 = 16.875
ohms, so  Z=16.875/(2*pi*F) = 0.0537H needed.

The MOT @ 53.4 mH was spot on and did indeed give me around 16 amps.

The coil ran well, but the safety gap fires like there's no tomorrow. (it's
setup correctly for the new tranny voltage).

Adding the welder ballast at 60.5mH = total of ~114mH and the problem still
occurs, but slightly less. Rather oddly the current though still seems
around 16 amps.


If I remove one string of caps and thus drop Cp from 43.2nF down to 32.4nF,
everything is fine, even with just one MOT at 53.4mH . 
No sparking on the safety, but the Cap value is extremely STR now, and no
doubt below what could be expected. I  really need to increase the original
43.2nF for the new tranny, not decrease it ! But I want to see what exactly
is causing this behaviour before I bother Dr R for some more CD's.


The archives gave some interesting posts on this but I can't seem to find a
definitive explanation as to how you work out what ballast value resonates
with Cp. 

Working out for the secondary side, I understand that if 1/(2*pi*SQR(L * C))
= 50 Hz then we have resonance, but to get the 'L' value what figure are we
using? 

The 'L' of the ballast alone multiplied by 'n' or the L of the ballast and
the L of the primary added and multiplied by 'n'.  In that case the MOT's
53mH + the primary's 176mH give a total of 229mH . This reflected to the
secondary equals 404H.  This would mean resonance at 1 / (2*pi*SQR(404H *
42nF)) = 38.64 Hz 

38Hz is a fair way away from 50Hz? Is that not enough ?


It also sparked with the welder AND the MOT.  This was 60mH + 53mH + 176mH
primary = 289mH = 510H reflected on the secondary side. This would give
resonance at 34Hz though, even further away from 50Hz.


This all assumes that resonance is the culprit of course. Could it be
inductive kick back maybe?


Trying to run the coil without ballast and using just the Variac, means it
won't fire at all unless you set the Variac at 270v and flick the switch a
couple of times - then wind the Variac back pretty darn quick as it is only
a 10 amp one!


On my 10K/150 NST setup and using all the Cp it was well behaved, allowing
you to get down to 80v input with around 0.5 second firing, so the coil is
not at fault.


So QUESTIONS at last if your still awake.


1) Is it resonance with the ballast that's the problem?


2) Am I working it out right. Do you add the L of the ballast to the L of
the primary and reflect both values over to the secondary ?


3) If not resonance problems is it inductive kickback?


4) Why the odd behaviour with the Variac alone?


Regards


Phil

www.follytowers.co.uk/tesla


_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla


_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla