[AMPS] parasitics
km1h@juno.com
km1h@juno.com
Sat, 25 Apr 1998 08:39:29 EDT
On Fri, 24 Apr 98 15:57:07 -0500 Jon Ogden <jono@webspun.com> writes:
>
>>
>>But a good suppressor should work at 50MHz also. If the natural
>parasitic
>>of a 3-500Z is somewhere around 130MHz why on earth would you use a
>>suppressor that can barely perform at 28MHz? There is no rule that
>you
>>have to design a suppressor based on the frequency of normal
>>operation.....you design the suppressor based upon the tube and then
>it
>>is stable wherever you want to use it.
>>Rich's comment about 6M being VHF is just a bunch of bluster since
>the
>>natural parasitic of the 3-500, 572B and most common ham tubes is
>well
>>above 50MHz.
>
>You bring a good point here. However, in the case for my amp, I don'
>think the orginal supressor I made would work.
>>
> It was down to ONE turn before it worked correctly.
Many published 4-1000A amp suppressors use 2-3T of 1/4" or larger strap
around a Globar.
Bill Orr also suggested using 2 or 3 2W carbons in parallel with the L
around only one of them. This goes back several years when Globars became
scarce and all the old junked tube TV sets were no longer available.
>>With #18 wire I dont doubt it. I have never seen a 4-1000A circuit
>using
>>such a ridiculous suppressor.
>
>Well, I was only following what I read in both the ARRL Handbook and
>W6SAI's book (including a c.1972 copy). It is amazing how unhelpful
>most
>of the literature is out there concerning amps.
That particular amp was pbly never tested on 10M Jon! They have the same
one in my 1981 edition but almost all the remaining amps use 3 resistors
and 1/2" wide strap.
In any case the GDO is a valuable tool.
>>
>>A properly engineered suppressor should work on either band. A
>typical
>>ham schlock suppressor is a gamble in either case.
>>
>So I guess I am a schlock designer! :-)
I never said that Jon...watch it before some senile citizen gets his nose
out of joint again!
>
>>Again you are completely missing the point. For extreme simplicity
>just
>>visualize the amp plate circuit as 2 distinct circuits. One on the
>left
>>side of the plate blocking capacitor and the other on the right side.
>
>
>ARRGH! Carl, no I am not missing any point. I couple my GDO to the
>tank
>coil...I see a resonance at say 14.250 MHz. I couple my GDO to the
>supressor inductor, I see a resonance at 85 MHz. Sure I am using
>different coils, but the entire circuit is still resonant at 85 MHz.
>This is with the tube in place. Not without it in place. If the
>circuit
>is resonant at 85 MHz, it is resonant at 85 MHz.
Of course...I fully agree and that is what the suppressor should be
designed for. It would also help if you could shift that frequency higher
with lower inductance in the plate strap. In some cases you can try
moving just the Suppressor R, no coil, along the plate strap and GDO for
best position. Other circuits bend the plate strap into a small U and put
the R across the open end.
Mouser has no minimum $ order so it would be advantageous to pick up a
handful of various value 5W metal oxide's.
I don't know why we
>can't agree with this.
I think we do agree; just communication confusion!
When a GDO shows a resonant dip in a circuit
>it
>is obviously resonant there as it sucks out energy from the GDO.
>Unless
>I am a totally dense dipwad, I don't understand what you are getting
>at.
>Airhead maybe (I'm blonde), dense dipwad, no.
Stop that...I can hear the bozo chorus warming up! :)
>
>>Yes, yes, in reality they interact BUT you do not design the
>suppressor
>>for what happens on the right side of the blocking cap.
>
>Depends on which side of the amp your tube is on. On the right side
>of
>my blocking cap is my supressor circuitry. On the left side is my
>tank
>circuit. So I am designing the supressor for the side of the circuit
>that includes: tube, plate RFC, supressors, blocking cap and tune C.
Huh? The Tune C is on the same side of the blocking cap as the tube????
>>Well, I guess we do. Rich is correct when he says that a magnetic
>>>field
>>>produces a mechanical force.
>>
>>
>>If you are quoting from a textbook then a resounding YES.
>
>Hmmm...You doubt that current flowing through a wire causes a magnetic
>
>field and that a magnetic field has a magnetic force?
More confusion again Jon. All I am asking is for you to explain how a
parasitic has enough energy to generate a field to bend a filament.
Nothing else.
>
>I don't need to quote a text book to do that.
>>
>>I will try AGAIN....explain only...not hyperbole...how a 3-500Z
>filament
>>can be bent. If you have to resort to EE math do it. What I cant
>follow
>>my son can.
>
>Now, imagine that same sort of magentic force except instead of a 1.5
>volt battery, you have a 6 Kilovolt supply with a capability of 2 amps
>of
>current. That's 12,000 Watts
A 3-500Z runs at 3KV at 400ma typically. How did we jump to 6KV and 2A?
The filament runs at 5V @ 15A and produces a magnetic field.
A parasitic will not generate enough energy to bend anything...BUT a gas
arc can. You seem to continually neglect the gas theory...why?
Did you know that mechanical power is
>also expressed in Watts? Hmm..could there be a reason.... Ok now
>hold
>that nail with 6 KV and 2 A of current next to a cage of fairly thin,
>but
>white hot wires. What do you think will happen? Do you think there
>will
>be a magnetic field around that wire? What happens to metal when it
>is
>white hot? From watching blacksmiths when I visited Plymouth
>Plantation
>as a kid, I learned that white or red hot metal is malleable and
>flexible. I'll leave you to figure out the answer.
Meaningless to the discussion IMO. I have no idea how malleable a white
hot tungsten filament is...do you?
>
>
>>
>>> I've seen lots of transistor amplifiers go
>>>*POOF*
>>>due to this happening.
>>
>>Sorry, but that example has absolutely ZERO meaning in a tube
>circuit.
>>
>
>Please tell me why? Tubes are very much like FETs. I fail to see why
>a
>failure mode in a solid state PA couldn't happen in a tube. The only
>thing is tubes ARE more rugged and can take more abuse.
Absolutely true. A FET junction will most likely be destroyed by a
parasitic. The parasitic frequency is also often at LF.
>>
>>I have zapped my share of SS stuff...it is a totally different
>scenario
>>but I can see that with really no tube experience you use that
>example.
>
>Quote from a textbook as to why.
Primarily since the SS device is a junction device and a tube requires a
vacuum as the conducting medium. I just dont think it is wise to lump
them both into one for the purpose of this discussion.
>>Yet you profess to accept a theory with no proof! That is strange
>>behavior for a trained EE.
>
>Carl, do you know what the "Scientific Method" is?
The rest SNIPPED since I don't have my hip boots on!
Goto go outside and play yard work..too nice a day to waste here!
73 Carl KM1H
_____________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com
Or call Juno at (800) 654-JUNO [654-5866]
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm