I have been dinkin' around to produce a high current, relatively high
voltage supply for use with the H-at- thyratrons and small Tesla coils.
Some tips follow:
1. Try to use only a current limited power transformer. I plan on using
a 60ma 15kv neon transformer. This auto-limit feature keeps big trouble
from happening in case of shorts or lock up in the H2 Thyratron.
2. I plan on using a variac for controlling the primary voltage to the
transformer so that I can make the supply output variable.
3. By ground referencing the center tap on the secondary and using two
diodes, I can achieve full wave rectification without a bridge. The
downside is the output will be limited to the half the peak voltage of
7.5KVAC (about10KV). This has a second advantage of making the minus
(negative) lead of the supply, ground. This can actually be a big
disadvantage if you ever want to float the supply. If floating is
demanded then you would have to bridge rectifiy the system and use a none
grounded, isolated secondary transformer. Such transformers are rarely
found in current limited models. I have one from an old heart
de-fibrulator unit by Hewlett-Packard (hamfest buy).
The diodes used should be fast recovery 1 amp units. I am going to use
series connected microwave oven diodes. I might use some special
unitrode 2 amp 7.5Kv diodes which I got at a hamfest. Some 673 mercury
vapor vacuum rectifiers would be the best of all worlds, but the filament
transformer and the isolation required on same is beyond most casual
coilers abilties. The resulting 50-200 watt coil system would be a real
weeny compared to the 200 watt filament requirement on this big 1
foot tall 15kv 5 amp glass diodes!
4. Metering is a must for both current and voltage. The current is best
metered in the negative or grounded leg of the supply or else the meter
becomes hot. The voltmeter is best placed in the circuit with its minus
lead to ground and the series resistors (use about three for standoff) to
hot. I am speaking of dedicated d'Arsonval meter movements here of
course. If the floating supply design is opted for then the meters
should be mounted in plastic and a secondary plastic face shield added
over them to protect the operator from elevated potentials within the
5. The next item is the filter capacitor. I am planning on using a
large 1uf 20KV unit which I have on hand. This can be a big ole nasty
mylar unit (Red glass caps), as they make great filter caps, or even some
old WWII paper units. If this cap is made too small, you will have no
reserve energy storage and the output impedance of the finished supply
will be too high for any serious work. I would say that the filter cap
should be on the order of 100 times the size of the Tesla resonator cap.
(my cap limit will be .01 ufd or even a good bit less). Note**** You
must supply a bleeder resistor across the capacitor terminals. I plan on
using a 20 megohm bleeder. (remember however, that at shutdown there is
a 60 second delay (3 RC) before you are relatively safe. I plan on having
a dropout relay with a 0.1megohm resistor in circuit so that when the
variac power is killed, the resistor is thrown across the output. Such a
relay will have to be custom designed or a vacuum antenna switch
6. I plan on placing formal current limiters in the system to allow the
supply to be tailored to a number of jobs. I will series a 20 megohm
resistor in series with the hot lead for a 500ua max output. I will also
be using a 200k ohm resitor (switchable) for a 50ma max output system.
The .5ma system can be used for relatively safe electrostatic
investigations, while the 50ma system can be used for small Tesla coil
systems and H2 thyratron experiments.
7. It is pivotal to secure a .5 henry or larger HV oil immersed choke if
you plan such a supply!! This "must have" component is only found at
hamfests. This choke must be placed in series with the output at the
last moment prior to the application. My choke will be a 1 henry 12KV
unit which I picked up at a recent hamfest. This choke will save your
8. Finally, the supply should be built on plastic or some other good
insulator and all components mounted on standoffs of some kind to avoid
internal supply arcing. I plan on using some 10KV silicon probe wire for
all interconects and silicon HV putty at every joint.
Such a supply would supply DC to a Tesla coil which would make an
incredible system with much more output for the buck than the simple AC
systems we are used to. Higher rep rates would be easily obtainable.
The usual warnings about shock hazards are still in place with the extra
precaution that you now have a 50 joule system capacitor (1uf -at- 10KV)
ready to kill at an instant in the high current mode. BE CAREFUL!!
I hope this helps those who have contimplated such a construction project
for that illusive DC coil system. If you need a schematic for this
simple supply, you might bone up on power supply theory prior to
building. You'll get it on your own. Those who are old time hams
(before the 2 meter free license weenies) or other electronic types
intuitively know what I have planned are ready to build now.
Richard Hull, TCBOR