[Amps] HF2500 Power Supply Caps
Jim Thomson
jim.thom at telus.net
Tue Apr 8 23:16:33 EDT 2014
### I bought a caseload of Buss HVU-3 fuses from SSON. $5.00 each..and
$3.50 in bulk.
5 inches long..and sand filled. They are easily rebuildable by removing the
ends. Remove
sand, replace the fuse wire..and pour sand back in. No rocket science
here.
The 5 inch long fuse is rated for 5 kv AC. I use em in both a hb
3CX-3000A7 amp..and also
used in am Eimac YC-243 amp.... socketless 3CX-6000A7. Both the 3x3 and
3x6 use a 225 watt rated grid.
## On smaller tubes, the 3A rated fuse wire is replaced with a 1A rated
fuse wire. The sand
or whatever is inside the glass tube will extinguish an arc very fast. Buss
also makes the same line
of fuses without the sand inside. You can tell the difference right away
by the weight per case.
## there is a huge difference in max sc kva rating of sand vs no sand HV
fuses. And yes, the 1A rated
HVU fuse will open up in 10 msecs with 4A flowing.....and less than 2
msecs with
50- 150A of fault current flowing.
## Buss also makes em in a 10 inch long format..rated for 10 kv AC. The
cheap and dirt way to make a
10 kv AC rated fuse is to use 2 x 5 kv fuses in series.
## The Eimac 3CX-3000A7 as used in the infamous 8 k uktra has been prone to
grid to cathode shorts
for a long time. I use the Svetlana 3CX-3000A7 ....which uses a spiral mesh
fil. I use a sola constant
V xfmr....running on a dedicated 240 vac line....to feed the input of a
0-240 vac at 5A varaic.....which in turn
feeds the dahl fil xfmr. A cam on the variac shaft + relay setup makes
for a fail safe circuit. If I forget
to turn down the variac..or the commercial AC power goes dead, there is no
way the fil xfmr will get max voltage.
The variac has to 1st be cranked fully CCW..til the cam engages. Relay
clicks on..then variac now has its 240 vac applied
to the input of the variac. Then variac is slowly cranked up as per
normal.
## Ok, now you wont get any grid to cathode shorts. Amp manufactures are
too cheap to use a 5 inch long Buss
HVU, 5 kv series rated HV fuse. The damn things are $35 each....min qty of
10....from mouser. Im sure ameritron, and alpha
could get em a lot less..when buying in large quantities. However, a HVU
fuse of the correct size + a 50 ohm glitch R
will suffice to do the job. If the tube arcs from the anode to the
grounded grid, the tube is toast anyway. Either it didn’t hi
- pot test high enough to begin with, or other issues. Just cuz Ameritron
et all are too cheap to use a HV fuse doesn’t mean the
concept is flawed. In some extreme rare cases, we have had the 2nd HV
fuse....located in sec of plate xfmr open up. In no case did
the fastest mag hydraulic breaker in the 240 line ever open b4 a HVU fuse.
I ordered a custom built airpax breaker for the big
253 lb dahl xfmr. They remove the fluid from the reservoir......to make
it......instant trip.
## per Buss charts.... 4A will open up a 1A rated HVU fuse in 10
msecs....and 5A will open it up in 1 msec. You can verify using a
test jig on the bench.....plus a scope.
## I have used the sand filled HVU series fuses mounted both horizontal
and also vertical......it works both ways. I use .75 inch
silver plated fuse clips mounted on ceramic standoffs. The fuse clips have
end retainers.....so the fuse can not migrate out either end.
Its captivated once installed. The .81 inch OD ends of the HVU fuses fit
good and tight in the .75 inch od fuse holders.
## If u want to add electronic grid and plate overcurrent protection....go
for it. I would myself, if using tubes with floozy grids like the
4 watt grid used in the Eimac 3CX-800A7.
Jim VE7RF
-----Original Message-----
From: Paul Hewitt
Sent: Tuesday, April 08, 2014 9:23 AM
To: 'Jim Thomson' ; amps at contesting.com
Subject: RE: [Amps] HF2500 Power Supply Caps
Melt time does not equal stoppage of current flow at these voltages. The
fuse element is replaced with a low impedance arc that keeps on pumping
energy into the fault for several AC cycles until the fault clears, mains
fuse blows or the voltage drops below the level to maintain the arc which is
much lower than the voltage that started it. The fastest fuses are very
long and filled with sand or other compounds, also pricey. I've seen the
binding post HB fuse and they have the same problem of slow time. If
someone thinks these methods are protecting a #40 grid wire in the Eimac
tubes most of us are using it's a dream. If the target is compliance with
Eimac's bulletin 17 then this doesn't do it. If better than nothing is the
goal then it's fine. If there really was a fuse that would stop current
flow in 2ms at these voltages every amp manufacturer on the planet would be
using it.
I pulled a 3cx3000a7 out of an 8K ultra yesterday with a dangling grid wire
causing an intermittent grid to cathode short, ouch. I can buy a lot of
protection for the $1500 the tube will cost but not my money.
BTW this topic was beat to death on this reflector several years ago, more
reading in the archives.
Vry 73
Paul
Paul Hewitt
WD7S Productions
QRO Homebrew components
http://home.earthlink.net/~wd7s/contents.htm
-----Original Message-----
From: Jim Thomson [mailto:jim.thom at telus.net]
Sent: Monday, April 07, 2014 1:34 AM
To: Paul Hewitt; amps at contesting.com
Subject: Re: [Amps] HF2500 Power Supply Caps
## Check out the Buss HVU series of HV fuses...then extrapolate the curves.
Then test on the bench
with a scope to verify that data. BTW, those fuses can easily be rebuilt.
I have also tried hb hv fuses,
encased in either teflon spagetti tubing or heat shrink.....suspended
between standoffs. Then crank up the
screws tight, in a CW fashion, so there is lots of tension in the fuse wire.
## For folks that want to use a 25 ohm glitch and say a 3 kv supply.. fault
current is 120 A. Stuff 120 A through
a 1 A rated fuse...and you will soon see it blows open pretty damn quick.
Even with a 50 ohm glitch and same 3 kv, the 60A fault current will still
open up a 1 A rated fuse very fast.
## You could put the scope across the cold end of the last resistor in a
HV multiplier string.
http://www.cooperindustries.com/content/dam/public/bussmann/Electrical/Resou
rces/product-datasheets-b/Bus_Ele_DS_6003_HV_Series.pdf
You can see that with just 4A flowing through a 1A rated HV fuse..it will
open up in 10 msecs. Now stuff 50-60-120 amps through the same
1A rated fuse..and you will measure a LOT less than 10 msecs.
Jim VE7RF
-----Original Message-----
From: Paul Hewitt
Sent: Sunday, April 06, 2014 2:21 PM
To: 'Jim Thomson' ; amps at contesting.com
Subject: RE: [Amps] HF2500 Power Supply Caps
Hi Jim
I see your posts about high voltage fuses disrupting current in 2ms so how
about a part number/manufacturer. I would like to see the data sheets I2T
curves. Most of these fuses open times are measured in seconds and are very
large.
Thanks &73,
Paul
Paul Hewitt
WD7S Productions
QRO Homebrew components
http://home.earthlink.net/~wd7s/contents.htm
-----Original Message-----
From: Amps [mailto:amps-bounces at contesting.com] On Behalf Of Jim Thomson
Sent: Saturday, April 05, 2014 1:29 PM
To: amps at contesting.com
Subject: [Amps] HF2500 Power Supply Caps
Date: Sat, 05 Apr 2014 08:41:38 -0700
From: Bill Turner <dezrat1242 at wildblue.net>
To: amps at contesting.com
Subject: Re: [Amps] HF2500 Power Supply Caps
On 4/5/2014 8:06 AM, Carl wrote:
> but more C doesnt buy anything useful.
REPLY:
More C can actually be harmful. In case of an arc, more C just dumps more
energy into the arc with potentially more damage.
73, Bill W6WRT
## With a 50 ohm glitch Resistor installed, the peak fault current remains
the same, regardless
of C used. However, the 50 ohm glitch + total C will form a RC time
constant, whose duration
will be longer when more C is used. With 2500 vdc + a 50 ohm glitch
Resistor used, peak fault
current is 50 Amps..... which will easily blow open a 1-2 Amp rated HV
fuse...and under 2 msecs.
## The 50 ohm glitch resistor LIMITS the peak fault current. The HV fuse,
just prior, and in series with the
50 ohm glitch resistor, INTERRUPTS the peak fault current. You can cro-bar
the supply all day long if you want.
All that happens is the HV fuse blows open.
## use whatever size cap you want. The 440-600 uf types work just fine.
The bigger the total C used, the better the
dynamic regulation of the supply. The ripple is also lower, and in
proportion to C used. A 600 uf cap will have one third the
ripple of a 200 uf cap..and way better dynamic regulation on ssb and cw.
Your load line is not all over the map.
Now this all works with the proviso that a 50 ohm glitch R and HV fuse is
used.
## As far as regulation goes, you require a plate xfmr that doesnt sag to
begin with....and big enough ga wire from HV supply back to main 200A panel.
You also require large enough rated relays and
contactors. Any Z between main panel and plate xfmr
will create V drops..which is bad news with any high C filter.
Jim VE7RF
_______________________________________________
Amps mailing list
Amps at contesting.com
http://lists.contesting.com/mailman/listinfo/amps
More information about the Amps
mailing list