[Amps] mallory caps for HV supply.

[Jim Thomson]

 The 1N5408 negative rail clamp is a good idea but JI vehemently opposed 
that idea every time I suggested it and claims the 1N4007 is better. 
Unfortunately Ive had his "expertly designed" amps in here several times for 
service with vaporized 1N4007's when a direct short is what they were 
supposed to do.

##  A  single  1N4007 is the wrong diode for the job. It's only a 30A surge rated device.
The 1N5408 has a 200A surge rating. The 6A10 has a 400A surge rating.   I use 2-4 x 
6A10's , all in parallel, all pointed the same  way. Then no more replacing safety diodes.
Use 6A10's across the meter's as well. 

##  I install a HV fuse just inboard of the 50 ohm, 50 watt  glitch resistor.  The 50 ohm glitch R
will LIMIT the peak fault current  to a safe value.  The HV fuse will  INTERRUPT the fault
current, and do so in < 2 msecs.  All that fault current has to pass through the safety diode
installed between chassis  and B- rail.   





Thanks to KM1H and K0UYA for the advice. The form factor of the stock
> caps for the Titan is the killer: very long and very narrow compared to
> the "usual" for this capacity and voltage. I'm confident I could make a
> safe outboard "capacitor box" bolted to the side of the current PS to
> allow more available/affordable capacitors, but I have to weigh all the
> factors, and $29 each at Newark may be a better choice. I don't know if
> 270uf/8 is sufficient vs 400uf/8 for an amp that can put out 1.5kw with
> the figurative pinky finger of one hand. So I'm sure I want to stay with
> 400uf per capacitor, while also paying close attention to temperature
> and rated life specs, ESR, etc. This is why I asked about a better deal
> for *the specific Mallory (CD) capacitors* before exploring unknown
> territory.

###  No big deal  using an outboard cap box.  I have done just that several times.
I have case-loads of  2300 uf @ 450 vdc lytics, all with 10-32 screw terminals.
I put a 100 k- 3 watt  mof across each cap. [1% tol, mouser item].  With 8-12
x caps in series,  you end up with loads of uf...like   230 uf  with 10 x caps....and
also a healthy 4.5 kv rating.    The resulting dynamic regulation on ssb/cw is nothing short
of phenomenal. Ripple is  reduced to essentially zero. Your pep output on the wattmeter
will increase a bit. 

##  with any  B+ to chassis arc...or B+  to grnded grid arc, still no big issue.  The peak fault
current is the same, regardless  of how much uf you have in the HV supply.   Peak fault
current is just  B+ / glitch resistor value.  3kv/50 ohms =  60 amps  fault current. 
60A  fault current through the B+ hv fuse [ rated at 1 A or whatever u require for normal
operation]   will blow open the hv fuse asap. 

##  the HV step star is no big deal either.  I used a 10-20 ohm metal cased arcol type
resistor in one leg of the 240 vac line.   It gets shunted out after a 5-10 secs.   I built a separate
step start box, consisting of a 30A DPST relay to apply the 240 vac.   And also a 30A SPST
2nd relay, that shunts the  step start resistor.  You can either use a 0-10 sec  timer to activate the
step start spst relay...OR... just use 2 x separate toggle switchs.... one for each relay. 

##  If you require any HV caps, I have caseloads of em, new, $12.00 per cap.  2.5" diam x 4" long,
14 oz each.   ESR is .026 ohm   Ripple current rating is 10 A  CCS.   We use the same caps in everything
from 3-500Z amps  up to 15 kw amps.   They come with the 10-32 machine screws + internal tooth lockwashers. 
If you need the 100 k 3 watt mof's [ these measure at .2% tol]  I have those too. 

later.... Jim   VE7RF 





>
> The Digikey price for these caps is $46 each, if they had them in stock,
> which they don't. Mouser does have 15 in stock at $81 each. Other
> suppliers I've found all over the planet have prices around $35 or more.
>
> Anyway, I wanted to make sure there isn't a golden supplier hiding out.
>
> Regards,
> Pete AD4L