Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

donate to eham
   Home   Help Search  
Pages: Prev 1 2 [3] 4 Next   Go Down
  Print  
Author Topic: Building 4-1K amp - parasitics and other problems  (Read 8267 times)
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #30 on: July 01, 2008, 04:23:02 PM »

<<
It's going to take me a while to digest all of this.
First off what would be an acceptable suppressor inductance and construction for this application?>>

The problem is every layout is different. Think about what is happening. At some frequency the grid connection inductance is parallel resonant with grid capacitance. In a tetrode you have TWO grids that might be a problem.

If the anode circuit is resonant near or above that frequency the tube can easily oscillate.

Not only does the anode path construction control the required anode inductance, so does the grid wiring.

At whatever frequency the grid lead is parallel resonant the anode path has to have an impedance that is either very low or mostly resistive. So obviously anything required is heavily layout dependent.

In a good layout with wide smooth short anode leads (like 3/8th inch strip or wider) and a vacuum cap mounted solidly to the sheet metal, and short wide grid leads through the socket slots to the chassis, it probably takes about five turns 3/4 inch diameter and 1 inch long. That's about 300nH and 200 ohms XL on 100MHz. The resistor would probably be 100 ohms or less so it dominates the circuit on 100MHz.

What is an acceptable resistor inductance? A lot of these metal oxide film resistors do have some spiraling in their construction.>>


Why don't you buy some Ohmite OY series resistors? Like these:

http://www.mouser.com/Search/ProductDetail.aspx?qs=%252bTRNBkpqqgisrj2sYdEIZg%3d%3d

They are metal composition styles. They are handy to have.

<<I don't have a lcr bridge but I couldn't get any of these resistors to measure any inductance on a cheapy meter I have or resonate with a 100pf cap using a grid dipper up to 250mhz. >>

I'm not sure you can grid dip the resistor because coupling would be low. It takes pretty high Q to dip something. The important thing is you want the current to shift over into the resistor fast above ten meters. A capacitor in the order of 50 or 100 pF in series with the resistor will help that a great deal, because while the inductance is increasing in reactance the capacitor will be dropping.

Look at what the AL-811H amplifier does.

<<My anode circuit is mostly 1/2 inch copper strap except for the 3 turns of #12 in the suppressor. The vacuum variable is mounted to a bracket that is bolted to the chassis as close to the tube as I can get it. I have a glass chimney. The loading variable is also bolted directly to the chassis not to the front panel. >>

Sounds good.

<<I guess in summary this all boils down to I need a tuned input, probably more drive and a better suppressor not to mention a better power supply.>>

I wouldn't do a thing until you add the tuned input and see how it acts. Change one thing at a time. First order is the tuned input with a lowpass configuration. Do it right near the tube.

To ensure high efficiency you have to have a low impedance cathode to chassis path on the second harmonic of the drive frequency. The capacitor on the tube side has to have a low impedance on the second harmonic, so for ten meters you would probably want to use a 50-100pF cap even if that results in an input Q of 5 or more.

You have to get rid of the second harmonic generated in the cathode by shorting it to ground.

All of this involves some cut and try because the layout has a large effect on values.

If you look at the AL1200, and that tube is the rough electrical equal of a 3-1000Z since it uses the 3-1000Z filament and grid, there are two 50 ohm coaxial lines in parallel to the tuned input. This 25 ohm line allows moving the tuned input away from the tube without hurting efficiency.

If that line is replaced with a single 50 ohm line efficiency on 15 and 10 meters drops about 10 % or more. This is a trick that allows you to move the input circuit away from the tube.

A pair of 3-500Z's I did for Heathkit went from 45% anode efficiency on ten meters to 63% just by adding a second line from the tubes to the tuned input. It was a ~17 inch long lead.

Have fun and be careful.

73 Tom


Logged
WB2WIK
Member

Posts: 21836




Ignore
« Reply #31 on: July 01, 2008, 06:57:13 PM »

>RE: Building 4-1K amp - parasitics and other probl  Reply  
by KD7YQM on July 1, 2008  Mail this to a friend!  
Steve: I'm pretty sure I'm not getting enough drive on 15m because the grid current is fairly low (50ma) at full output of my driver (old ft-101b). I'm probably not getting enough drive on other bands or I probably would have more output. I do not think my power supply is capable of 700ma anyway. I think I do need that tuned input circuit. Someone please give me a good example.<

::Gosh, you're not using a tuned input with a grounded-grid circuit?  For sure that's a mistake almost always.

My 4-1000A amp (c. 1987) is not grounded grid, it's grid driven and grounded cathode.  50mA would be a huge amount of grid current for it, since it will run 1500W output without showing any grid current at all while speaking (there is a little, it's just very small, like 2-3 mA peak).  4kV on the anode, 660V on the screen and the plate supply can provide 1 A continuously (designed for RTTY or other key-down modes, uses Dahl hypersil xfmr rated 1.2A CCS and weighing 60 lbs -- just the transformer).  The power supply in entirety weighs just shy of 170 lbs.  At 750mA Ip, plate voltage drops about 120V.

I made it grid driven because I didn't know how much drive power I'd have available with these "new fangled" solid state rigs.  I was used to tubes and just transitioning to SS transceivers.  My TR-7 could very seriously overdrive this amp, which is why the amp has a 6 dB/150W resistive attenuator right at its input port on the grid side of the T-R relay, and I use the ALC connection to keep the exciter under control (to a point).  I'm using an E.F. Johnson ceramic socket.

I built another 4-1000A for 50 MHz only and it's stable as a rock but of course only one band.  That amp was on the cover of CQ magazine back in '85 and the subject of a construction article.  I know several others who built it, and they're all stable.  Runs 1500W PEP output with 22W PEP drive power.  Reason for that is I wanted to use it with a SS transverter that only delivered 20W PEP output.

Sounds like Tom's built a whole lot more 4-1000A amps than I have, but my experiences were "fun."  I found I could cut winding grooves into a 1" Teflon rod, for example, to make RF chokes, using an end mill.  And I found that "TV" high voltage doorknobs from CRT anode supplies can suck at HF, no matter their rating!

WB2WIK/6

 
Logged
KA5N
Member

Posts: 4380




Ignore
« Reply #32 on: July 02, 2008, 05:58:22 AM »

<"Do you know if these pi circuits will be broad enough so you don't have to tune anything?

My driver has a tube output that can tune 25-100 ohms or something like that.">

Well one designs for a low Q which means that the bandwidth is broad.  If one uses standard silver mica capacitors the tolerance is +/- 5%.  If you use powdered iron toroids you will have an additional tolerance.  If you use variable caps then you can tune to the middle (or bottom or top whatever you desire) of a band.  With a driver with a pi network output you can tune to compensate for small frequency offsets.  What you are aiming for is a low pass filter which matches 50 Ohms to 105 Ohms.
It would be to your advantage to study filters, circuit Q, etc. before starting to build them.  A good understanding of circuits allows one to correct a circuit that doesn't behave properly.  As you can see from the length of this thread how difficult it is to build something and THEN design it.
Allen
Logged
KD7YQM
Member

Posts: 71




Ignore
« Reply #33 on: July 02, 2008, 09:13:43 AM »

I tried the low pass filter on 15m. I think things are worse. I think I got even less drive than before.
I used a powered iron core (amidon mix 6) 1 inch diameter. .498uh with 4 turns. Source cap is 176pf and load cap is 142pf. I used nearest value silver mica.

The suppressor resistors got very hot. I shut down before they burnt. Must be an incredible amount of power across that 3 turns of #12 wire. It's got to be vhf to have that much power across that coil.

This tube really wants to oscillate at vhf. I guess I'm going to have to take toms advice and get those screen grounds shorter.
Logged
KD7YQM
Member

Posts: 71




Ignore
« Reply #34 on: July 02, 2008, 01:47:42 PM »

Tom
I reworked the screen leads like you said. Got the straps through the socket slots and grounded to the chassis.
I'll be doing some tests to see if there is improvement.
Dennis
Logged
KD7YQM
Member

Posts: 71




Ignore
« Reply #35 on: July 02, 2008, 02:09:17 PM »

Just ran a test with 15m. I am now getting the same output on 15m that I got on 20m. And that's without burning the suppressor resistors.
This is after shortening the screen leads as per Tom's instructions.
I still need to get up the nerve to try 10m. But I think there is hope. ;>)
Logged
KA5N
Member

Posts: 4380




Ignore
« Reply #36 on: July 02, 2008, 03:32:20 PM »

Sounds like there may be light at the end of the tunnel.  I sure didn't think a tuned input would make things worse.
Good Luck
Allen
Logged
KD7YQM
Member

Posts: 71




Ignore
« Reply #37 on: July 02, 2008, 03:41:22 PM »

Hi Allen
Yes now that I may have the self oscillating cleared up I might try the tuned input again.
Logged
KD7YQM
Member

Posts: 71




Ignore
« Reply #38 on: July 03, 2008, 07:40:42 AM »

I have bad news. Unfortunately the problem is not gone.

I powered the amp up today and was getting full output on 15m, then I noticed that the suppressor resistors had melted their solder connection and were not even connected.
At this point I measured them and they are total open circuit.
I replaced the resistors with new ones. Now I'm back to half the output power again??? The resistors started to burn again.
I also noticed that the swr was rather high between the amp and the dummy load. ( swr is fine on 20m)
Just for curiosity sake I measured the inductance of the suppressor choke. It measured .32uh.
Anyone have a clue as to what's going on?
I know this thread is getting rather long and everybody's getting a bit frustrated by now. I don't know where to turn at this point.
Logged
K2XT
Member

Posts: 39




Ignore
« Reply #39 on: July 03, 2008, 05:38:39 PM »

The SWR to the dummy load is high?Huh   What ? You sure don't need any advice from any of us on THAT issue.  Some things we have to take for granted and a 50 ohm load that has good connectors and a good cable is one of them, right?  All bets as to stability are off until.
K2XT
Logged
KD7YQM
Member

Posts: 71




Ignore
« Reply #40 on: July 03, 2008, 05:59:18 PM »

There is nothing wrong with my load. It's a 1500w bird and reads exactly 50 ohms. Connectors are all perfect.
I read 0 reflected power with the driver alone. Very small reflected power with other amps connected.
This is a low frequency load so something in the vhf range would cause reflected power.
Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #41 on: July 04, 2008, 04:50:29 AM »

You are working with one of the most difficult tubes around to stabilize.

You could have considerable fundamental RF current through the resistors.

Change one thing at a time, and stabilize the amp before trying another major change.

First key the amp with no drive BUT with the exciter and load connected. Now watch the plate current and grid current carefully as you tune the plate tuning capacitor. Does anything change as you move the plate tuning cap through its range with NO drive but the amp keyed?Huh

Run the maximum expected HV and remove any cathode bias.

You will clearly see a change at some positions of the plate tuning cap on some of the higher bands if your amplifier has any stability issues.

Is it stable?

73 Tom



Logged
KD7YQM
Member

Posts: 71




Ignore
« Reply #42 on: July 04, 2008, 08:09:33 AM »

Hi Tom
I'm thinking the same thing. Remember you said you thought the parasitic choke might be too big?
I went from .3uh to almost 1.0uh and the resistor problem got markedly worse.
Right now I have almost 20w worth of resistors over the choke and it's getting very hot.
I'm going back to .3uh or maybe smaller. It goes to reason that the more reactance you have in that choke the more of the fundemental you will see cross the resistors.

I'm not seeing any effects of unstability now. I did before I shortened the screen leads. Although I an a bit perplexed by the small amount of reflected power I'm seeing.
I'm going to run it through it's paces with the bias removed and see what happens.
Shortening those screen leads I think helped a lot.

Dennis
Logged
WA1RNE
Member

Posts: 1010




Ignore
« Reply #43 on: July 04, 2008, 06:13:06 PM »

 
 Sounds like you need to consider another look at the suppressor design using low Q conductors - as well as a check of the output circuit for resonances at VHF.

 
 As one example, think about the effect of a few hundred nanohenries in the plate bypass capacitor leads - about the same inductance of your original plate suppressor inductor....not a good thing at VHF.

 How long are these leads and is the ground return as short and direct to the chassis as it could be?


....WA1RNE

 




Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #44 on: July 04, 2008, 08:39:48 PM »

The plate bypass leads have nothing to do with VHF. They are on the wrong side of the plate choke.

Run the test to see if you have a stability problem.

If you want to reduce resistor heating from fundamental RF add a small value mica or ceramic cap of a few kV rating in series with the resistors. Either that or reduce your inductor value so long as that does not affect stability.

You might also want to improve the grounding of the control grid.  You might be able to do that with a low inductance style capacitor instead of a disc.

Why did you put a resistor to ground there, instead of floating the negative lead and placing the meter between the negative lead and the chassis?? Then you could ground the control grid.

73 Tom
Logged
Pages: Prev 1 2 [3] 4 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!