[ARC5] Selenium rectifiers

[Charles]

> Well, everyone here can, of course, do what they wish with regard to selenium rectifiers, but I
> won't have them in the shack if I can avoid it.
>
> Most will deteriorate over a long period of time, some even while being stored. They become,
> essentially, resistors, not rectifiers.
>
> Then, if they short out under load, they spit out a gas that is poisonous, besides stinking to
> high heaven.
>
> IMHO, it is far less hassle to simply replace them with silicon. than to continue to worry about
> them...and I am NOT in favor of shot-gun replacing any other component without testing it
> first.
>
> I can't and won't trust the darned things.
>
> In the unit in question, we were told that the selenium rectifiers in this rotary converter are in
> the voltage-regulation part of that unit. By the very fact that those are in that part of the circuit,
> I would be replacing them even if they "test" good....for now.
>
> Although I am not "intimately" knowledgeable of that particular circuit, it seems to me that if
> they control the voltage regulation of the unit, it would be far better to replace them, now, with
> silicon rather than depend on such an iffy component.
>
> It also seems to me that if that voltage-regulator was not completely up to specifications, that
> COULD be the reason for the unusually high current the unit exhibits at no load. If there was
> more AC on the output of the selenium rectifiers than there should be, for instance.
>
> Of course, you MAY have to consider and account for the differing voltage-drops of selenium
> vs silicon, but in most cases that won't matter.
>
> But since it is your equipment, you can do as you want.
>
> Ken W7EKB

Thanks for making it clear, not once but twice, that I can do as I want ;)

The regulator circuit controls both frequency (by setting motor RPM) and 
voltage (sets the generator field current). I've attached a copy of an 
almost identical schematic for a 1518-1-E. the only one I could find 
online. My 1518-1-H regulator does not have the "Frequency Regulator 
Relay" but is otherwise the same. Its schematic is glued inside the 
circuit box cover and is quite small so I'd have to take a picture to 
post it.

Note that phase A (115 volt 400 Hz output) is connected to both selenium 
bridge rectifiers. The frequency regulator on the left has a 
parallel-resonant LC circuit tuned to 400 Hz, so it will be very high 
impedance when the frequency is correct, causing the output of the 
bridge (which supplies the carbon pile coil current) to be low. The 
voltage regulator bridge has a variable resistance; coarse adjustment 
with a 12 watt Dividohm (slider) and fine adjustment by a small 
wirewound pot to set the output voltage, also driving a carbon pile coil 
from the output of its bridge.

Both appear to be working correctly, as I said. If the bridges were not 
rectifying properly, somehow making the field current of either the 
motor or the generator grossly high, either the RPM/frequency or the 
output voltage would be way off. They are not...

Yes, I could replace them with modern encapsulated silicon bridges. But 
I'd also have to add resistance to keep the coil currents as designed, 
and I don't know how much. I would be concerned about blowing the coils 
and the carbon piles are unobtainium.

Right now I'm building my BFPS onto a rack-mount panel and chassis, so 
further experiments with the inverter will have to wait a bit.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Bendix 1518-1 schematic.png
Type: image/png
Size: 373694 bytes
Desc: not available
URL: <http://mailman.qth.net/pipermail/arc5/attachments/20220312/2947ca82/attachment-0001.png>