HV Pulse Transformer Design Using Small Toroids

In gaped cores the flux goes to nearly zero when the current through the transformer goes to zero.
In ungaped cores the flux remains at it value when the current through the transformer goes to zero.
The preceding means that the pulse transformer and it's circuit should be designed to work together.

One reference (if you can get it) is
HIGH POWER-HIGH VOLTAGE PULSE TRANSFORMER
DESIGN CRITERIA AND DATA
FINAL REPORT
(PART I of TWO PARTS)
Covering the Period
June 1, 1950 through November 30, 1952
Prepared by:
P. Fenoglio
C. W. Peck
F. R. Richardson
H. W. Lord
A. Boyajian
February 1, 1953

Thanks Carl! I'll start looking.

Do you know what the periodical / publication this was presented in?

I really do not know how anyone can help you . . without you first outlining what it is is you are doing . . . what you have . . . and what you need . . Oleh

The report was under
CONTRACT NO. DA36-039-sc-117
DEPARTMENT OF THE ARMY PROJECT NO. 3-26-00-600
SIGNAL CORPS PROJECT NO. 32-2006-31

G. E. REQUISITION NO. NWK-60F200

ASTIA Document 28714

This document is through. It may contain more information than is required for a simple design.

This is being normally done in lighting applications. There are instances of generating as high as 25 to 60 KV with specified pulse width.Hence you need to specify pulse height in terms of voltage & current and pulse width. Air-gap helps to keep the size smaller. If you do not wish to provide air-gap for any reason, the core size becomes big- but it is not difficult to achieve / compute. No need to give out those defence project references etc.

No need to give out those defence project references etc.

I agree. However do you know of any other good reference.

Thanks for the information.

The application is for a HID lamp lighting circuit. I have a packaging contraint so the smaller the better. The general specifications are >18kv for at least a 50ns period. The peak I am getting with a 20mm 5000u upgapped Epcos toroid core is 24KV. I've calculated durning strike, I'm well into saturation for about 10ns. But the funny thing is my Vout matchs my Vin*turns ratio. Question, With such a fast pulse, are there still concerns for stresses or other negative core effects? Estimated lamp starts per product life at 5000, with 3 or 4 strikes per day. (not a military or automotive application)

With a nongapped toroid, my windings ratios are from 5 to 7 for the primary Vin I'm using. For a gapped, I have to use a 25mm core and at least a 12 turns ratio, which makes it larger and harder to manufacture.

Any thoughts are appreciated.