>> However Ive never seen a simultaneous blown choke and grid cap in the
>> decades Ive worked with the SB-220 series or any other amp using a
>> similar design.
>
> I certainly have.
>
>> Early production SB-220 grid caps of the postage stamp variety will split
>> at the seams from age and moisture ingress but the dipped micas seem to
>> last forever.
>
> ....and you know it is age and moisture, and not an arc?
The same way you know its an arc. Neither of us was watching when it
happened. In my cases there was absolutely no other damage and the case was
split at the seam, the weak link. I also have replaced many similar micas in
old receivers where even you couldnt claim were arc related. Since a large
number came from antique car radios the moisture possibility is most valid.
As with the SB220 the caps arent shorted, they exhibit leakage as with the
old style paper style. Some SB220 caps dont even show leakage when split,
they are all simply replaced as a precaution.
>
> I've got boxes of 50 year old micas here of the same style that aren't
> split from moisture.
So do I and many go back to WW2. Its a meaningless statement.
>
>> OTOH, The Ameritron 2 x 3-500Z AL-82 and the AL-80 with directly grounded
>> grids have reputations of taking out a tube and plate choke when there is
>> an arc.
>
> ...and you somehow know the arc takes out the tube, and not that the bad
> tube makes an arc?
> How would you know that?
The same way you seem to believe it is the other way around. Do you have
documented proof of the failure mode?
>
> The FACT is the HV in the AL82 is 3600 volts or so, while other amps are
> down around 3000 or less. This is why very rarely in AL80's do you see
> damage.
> When the AL82 was designed 3-500Z's almost never arced. The Eimac's were
> great. This isn't the case today with the foreign tubes.
Strange comment since I was seing the problem as soon as they hit the
market. With HRO being in the next town I often received calls to try and
fix without having to return to the factory.
Next you will say that the Eimacs werent great that week, month or year.
We all know about the Eimac reliability problems over some production runs,
Ive left visually internally damaged tubes out of the equation.
>
> Truthfully if the AL82 was being done today it would have a 20 ohm or so
> limiting resistor. Why 20 ohms? Because the ESR of the filter caps and
> choke combination is already ten ohms.That limits surge to 360 amps or
> less, since the arc path itself has some impedance. Adding ten ohms would
> make it 180 amps or less, but adding 20 would get it down to 120 amps or
> less.
So why hasnt a tech bulletin been released and a resistor added to current
production? It seems a dumb and irresponsible thing to not do it.
>
> As for the idea a grid choke limits current, let's go back to basic Ohm's
> law. When the anode faults to the grid from gas, a plasma forms in the
> tube. See:
>
> http://www.w8ji.com/fault_protection.htm
>
> There are only two ways the plasma is extinguished and the fault current
> stops:
>
> 1.) The HV is lowered until the current flow stops.
It is often called popping the breakers.
> 2.) The grid and cathode rise to the anode voltage, so the current stops.
>
> NOTHING else will magically stop the current.
>
> If we assume nothing in the anode lets go, then the only other option is
> the grid must rise to anode potential. When it does that the little 500V
> rated mica caps have 3000 volts on them, and they arc.
With 3000V at that massive instaneous current the capacitors wouldnt arc,
they would explode. The real truth with the early caps is that its usually
one or at most two out of the six that has a hairline crack. I'll repeat
that the later dipped micas do not short or explode.
Now the grids are grounded
> again despite the choke opening. The capacitors actually try to protect
> the chokes.
>
> If the grid actually has a successful choke opening and the little caps
> magically can hold off six times their rated voltage, then the grid rises
> to near anode voltage. Now the plasma simply passes current through the
> grid to the cathode system.
>
> It's almost silly to consider that some sort of fusing system.
Its even sillier for you to argue in face of real world facts. The chokes
are most often wired very close to the chassis or the socket mounting nut.
The pi's slam together and become one slab of copper. This then arcs to the
chassis/nut, the arc extinguishes and the user just stained his pants.
The next time you examine an zapped SB220 look for the arc marks under the
choke.
Sometimes it pays to stop relying on pure theory and let your eyes do the
job. You may learn something without such a stiff stance.
>
> The only proper place to add a fuse or series resistance is the anode.
>
>>And that is supposed to be designed to perfection.
>
> No, that would be the National. :-) I don't know why they still aren't
> around since a perfect product would certainly have a longer sales life
> than the marginally designed AL-12 series, which only have run 25 years
> now.
Obviously you dont know your history. The NCL-2000 ham sales were killed by
the FCC when their regulations outlawed amps with 20dB of stage gain in
order to combat CB use. I forget the actual limit dB they imposed. The amp
was still sold into the 90's for industrial, military and foreign use. When
the IRS auctioned off the remaining assets of National I purchased partially
built amps along with boxes of parts and completed the assembly. Some of
them have recently bought big prices on Ebay from collectors.
>
> On the other hand the AL82 and virtually all other amps on the market
> today or in the past were designed to give the best value for dollar of
> cost. Not a single penny was spent that wouldn't benefit the customer for
> the target goal. Most engineers understand it is very difficult to build
> something that is priced to dominate a market but doesn't cause too many
> problems.
Some companies and engineers stress price and volume over quality. Others
stress quality or engineering innovation and sell to those who can afford
it. I know where Ameritron stands on that chart.
>
> One of the very best amps on the market was the SB220, it had everything
> on the edge of being almost too cheap. When the power limit was raised to
> 1500 watts output, the whole game changed.
I wonder why a well designed amp with three less stressed 3-500Z's hasnt
emerged? The LK-550 was an attempt that missed.
Carl
KM1H
>
> 73 Tom
>
>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|