>Hi all
>
>I have previously posted a mail to this reflector concerning
>poor efficiency of an Ameritron AL1500 amp. (8877), But after
>all discussions on parasitics I decided to make some additional
>measurements, to see if maybe this could be the cause to my problems.
>The "poor efficient" was must pronounced on the higher bands
>(20, 15 and 10m).
>
>First let me say that I have experienced no arcs the tank circuit
>in either bandswitch or capacitors.
>
>The AL1500 uses a 8877 that runs in GG. Ameritron does NOT use any
>parasitic suppressor in the anode. Rich put my attention on this
>after my first mail, and I made up my own suppressor little do I know
>about it, just looking how others did.
It is my opinion that a VHF suppressor should not be randomly copied. A
suppressor should be Designed so that VHF current divides semi-equally
through the resistor (Rs) and inductor (Ls) at the primary
anode-resonance frequency. For example, if a 100-ohm Rs is used, Ls
should have roughly 100-ohms of reactance at the anode-resonant freq.
More ohms in Rs and Ls is generally better since it reduces the VHF-Rp
(parallel equiv. R) of the suppressor, which decreases the VHF voltage
amplification of the tube. However, more Ls means exponentially more
suppressor dissipation at 29MHz, so the designer must be careful to avoid
destroying Rs.
> I've got somewhat wiser during
>the past few months after reading the numerous number of mails
>concerning the topic.
>
>Back to the measurements
>
>I checked for resonance with a GDO coupled between the anode and DC-blocking
>capacitor and found several (very sharp) from 100MHz and up!
>I then made a test setup with the AL1500 driven by a TS-680 and connected
>to a dummy load and used a spectrum analyzer to "sense" the RF.
Since the tank L tends to act like a VHF-choke, a better place to sense
VHF RF is 10-20cm from the anode, using a1cm probe.
>Would have used a directional coupler to get exact values, but the one
>available covered from 0.1-2GHz, and I was short of power attenuators.
>My RF probe consisted of 10cm wire. I terminated the end of the cable
>at the probe end with a 6db attenuator to reduce reflections.
>
>I tuned up on 20m, and was I surprised.
birthday surprise, ok. . amplifier surprise, not ok.
> The AL1500 produced signals up to
>350MHz!!! No wonder the neighbor had complained about TVI. Parasitics or
>not, I don't know. The broad band spectrum only exists when the amp. is
>excited. Keying the bias on/off can't produce any unwanted behavior.
>And yes, I checked the TS-680 for harmonic suppression, and the signals
>was definitely not coming from the driver, though it wasn't the best I've
>seen.
>15m and 20m are the worst bands, 10m nearly the same. 40m better
>and 80m only random oscillations at high speed keying (tuning pulser).
>
>I removed the HV connection and connection to the DC-blocking capacitor.
>Inserted a little VHF ferrite suppressor between the tube and DC-blocking
>capacitor. Resonance around 100-120MHz were gone however I could still
>see some resonance at app. 140-150MHz using the GDO.
>I then removed all connections to the anode and resonance at 140-150MHz
>was still present!.
>
The 8877 will always present 11pF to the anode circuit's L, so the VHF
resonance can not be made to go away.
>The suppressor I made was a copper strap 6mm wide 3 turns OD. of 10mm and
>a 22 ohm 3-5W resistor in parallel. I know this is far off!!!.
>Despite the "high-Q" suppressor, I could see that the 22 ohm resistor had
>been VERY warm, however it was still 22 ohms.
>
Making Ls out of a more resistive material than Cu strap will decrease Rp
Without having to increase the dissipation burden in Rs by increasing Ls.
. Very warm is very ok because it means that VHF-Rp is low.
>Looking at the grid circuit, Ameritron have made the input filter in
>a separate box away from the grid/cathode compartment. The input filter
>(pi) is connected to the cathode with a coax cable. Could this cable
>cause any problems since it will act as a 1/4 wave shorted stub at VHF!
>
Yes. This resonance is easy to see with a dipmeter. Connecting a 10-ohm
R in series with 30pF, and connecting this assembly directly from the
cathode to gnd will help to dampen the resonance.
>Now I would like to cure this problem. I need some directions to design
>the suppressor for the anode and to put some damping resistors in the
>grid/cathode. Would also appreciate comments from owners of AL1500.
>
>73 de Soeren, OZ1FTU
>
It is my opinion that the grid should be grounded through the Eimac
grid-grounding collet, and that one needs to limit anode-supply fault
current to under 200a peak.. .
- later, Soeren
cheers
Rich...
R. L. Measures, 805-386-3734, AG6K
--
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
|