Split stator or single cap on tuner?

[ke7trp]

I have a few roller inductors and good supply of Air caps now. I am thinking of making a coax to coax tuner and hanging it on the wall next to the GK500.  This tuner will tune the Vertical flat. Its probalbly not needed but it would be nice to stretch the 160 portion up to 1945. 

After reading the orr book and a few other books on designs, I am wondering if I should use a split stator cap on the input as the Orr book suggests or if I should just wire across the split and make it one large cap.

The caps I have are 300 Johnson split stator. One of them is huge and the other is smaller and currently wired across.

Pros and cons?  The book suggests that having the split on the input has less harmonics... It does not go into this further.

Then, I found a design on the net that has a split on the output which confused me even more..

C

[KA8WTK]

What design are you looking to build? L-network, Pi or some other configuration?

[ke7trp]

http://www.raes.ab.ca/AntennaTips/AntennaTips.htm

The first schematic is what I set off to wire up.  Then I saw the second version with the split stator cap. That second design is in the orr handbook and that is where I saw it.

Then, I saw the last design showing the split on the ouput..

I want to use this tuner with coax. 

I want to use it down to 160 and up to 20 meters since I use the king on all bands for AM phone.  Its going to look real nice hanging on the wall in the corner next to the king.

[K1JJ]

Clark,

I use the first schematic, a "T"  network to feed the 50 ohms input to my linear and 4X1 rig from the 50 ohm exciter.  It works very well on the various bands.  50 ohms to 50 ohms is not an easy thing to match sometimes, especially if both are close to 50 ohms to begin with.  Be willing to try all three configs if the first try doesn't match perfectly.

You mentioned harmonics. That should not be an issue if your rig is using a pi-network. Most guys feed coax-fed antennas without a tuner and have no problems with harmonics.

T

[WA1GFZ]

The only advantage of a split stator cap is no RF current flows through the rotor to ground joint. Also in a tuner, the case of the cap does not need to float so shaft is at ground.

[ke7trp]

Ok.. Sounds Like I should just strap both split stator caps up and run like the first diagram. I will try it.. I am bored anyways right now.

C

[k4kyv]

The only advantage of a split stator cap is no RF current flows through the rotor to ground joint. Also in a tuner, the case of the cap does not need to float so shaft is at ground.

If the split stator is used in a balanced tuner, you want to leave the rotor/frame/shaft floating.  It should already be at or near rf ground potential if the circuit is anywhere near balanced. Grounding the frame invites common mode currents, as the grounded frame serves as the ground connection of a Marconi antenna formed by the combination of feed line and flat-top.

In the Ultimate Transmatch and the W1FB tuner as shown in the above link, the split stator cap serves as a capacitive voltage divider.  This is not a balanced tuner. I never could understand the popularity of these circuits, since they provide nothing more than a 2:1 step-up or step-down using a capacitive voltage divider. A better method would be to use two separate capacitors, individually adjustable, or else modify the split stator into a differential capacitor, so that the step-up/down ratio could be varied. I would prefer to save my expensive split stator capacitors for use in balanced circuits.

[KA8WTK]

If I read the first schematic right, you will want to use an insulated shaft and coupler on both input and output cap. Neither stator or rotor are grounded.

Bill

[Steve - K4HX]

Read Walt Maxwell on why the split cap in the Ultimate Transmatch is a waste.

A reversible L or T would be much more useful.

[ke7trp]

Ok.  I got the thing built guys. 

Two big Johnson Split caps on a board.  The Roller in the center.  I made a mistake at first by not realizing that the curved line on the cap symbol was the section of the caps that moves. I had them backwards. This worked but when I touched the knobs, the tuning would change.

Once fixed so the Frame of the caps are hot to the coax jacks and the fixed plates go to the coil, I can touch the knobs with no change.

The tuner works FB. I can tune 1.0/50 ohms J0 on 160 and 75. I have not tested it higher. 

The only thing that is bugging me now is that if I touch the plastic knob on the roller, it throughs the tuning way off. The only way to tune it is to turn the knob, Pull your hand back a bit, Take a reading and then turn the knob again. 

Another question:

On the roller you have three connections..

1. North end of coil.
2. South end of coil.
3. Slider tap point that moves when you crank the knob

1 is not hooked to anything.
2 is hooked to the frame of each cap just like in the shematic.
3. is hooked to Ground or coax shield.

Should the oposite side of the coil be open or grounded???   Some schematics show this and others do not.  I put a jumper on it to ground and see no change. To me, Its just unused coil floating there... 

C

[Steve - K4HX]

The north and south ends should be connected as in and out. The slider should be shorted to one of the two ends. Get as shaft decoupler for the coil or use an insulated shaft.

[ke7trp]

That makes no sense.  One end of the coil is connected to the hot side of the RF.  The Slider or tap point goes to ground. This way, When you turn the crank you are making the coil longer or shorter to ground.

I grounded the unused portion of coil and there seems to be no difference in tuning.  I guess I will leave it grounded.

C

[KA8WTK]

If you are still on schematic #1, one end of the coil goes to the RF line between the caps. The other end and the slider go to ground. As the slider approaches the grounded end, the coil is "longer" (more inductance). You do need to put an insulated shaft on the coil so that your body capacitance does not change the tuning.

On some rollers I have seen the slider is already connected to one end and that end goes to ground. Also usually with this setup, the slider and the shaft will be at ground potential and not "hot". But, you may still need some insulated shaft to keep the proximity of your hand from changing the tuning.

[Steve - K4HX]

Oops, I thought it was a series connection. Bill is correct.

[ke7trp]

Thanks Bill. Thats the way I have it. I see other schematics shown with the other side of the coil grounded.  I ran a jumper there and saw no change in tuning so I left it.  Its unused open coil.I dont think it matters if its grounded or not.

The tuner works FB.  

160, 75, 40, 20 flat. The pupose of this thing is to stretch the 160 vertical. Its usable from 1810 to 1890 but if you try to go up past 1900, It gets out of hand and the King gets unhappy. Now on 1945, The king is happy as can be. The antenna might not be.. But at least I have 1/2 inch hardline on the run to the vertical.

Now to test it on the air.  I am going to transmit on 1885 with no tuner, Then put the tuner in line and transmit again to my friend.  Next I will move to 1945 and transmit to him and we can see if there is a huge loss in the tuner or in forcing the antenna out of resonance.

Thinking of making a cover for it. Would you worry about RF being in the shack with this mounted on the wall?  If so, I can make a Rabbit wire cover for it and ground it.

C

[KA8WTK]

Just remember, the T network, unlike the L network, has more than one combination of settings that will show you a good match at the input. Some of them will have high losses. Try you hand with this simulator to see what I mean.

http://fermi.la.asu.edu/w9cf/tuner/tuner.html

Glad you got it working!

Bill

[ke7trp]

Here she is guys:

http://www.youtube.com/watch?v=WZKQ4US4zvw

C

[ke7trp]

That is a really cool page.  I typed in my values and looking over at the position of the caps, I think I am close to optimal. The MJF is showing 1.0, 50, J0 and the transmitters plate current is exactly the same on the Bird dummy and the tuner.  I will have to run it for a while to see if anything gets warm.

What should I use to oil these oil caps? They are really dry spinning. 

Thanks again! I really learned alot today.

C

Just remember, the T network, unlike the L network, has more than one combination of settings that will show you a good match at the input. Some of them will have high losses. Try you hand with this simulator to see what I mean.

http://fermi.la.asu.edu/w9cf/tuner/tuner.html

Glad you got it working!

Bill

[K1JJ]

Nice job!!!!   Strapping look.  At low 50 ohm impedance, those caps could handle 10KW+...

Even though the TINY loss in the hardline will be the same, there's nothing like having a 1:1 50 ohm match at the transmitter's output.

T

[KA8WTK]

Looks heavy-duty!
You just gotta love the chrome "H" on the PM-23.

[ke7trp]

Thanks guys.. its a beast for sure.. If nothing more it will look damn cool mounted on the wall!!

The king is fine at 2 to 1. But past that, I got some spitting out of the Pi net.  Its low duty/ small parts in the king. Rather that mod the rig, I figured I would use this to keep her happy.

I guess I could also use this for a long wire as well. 

C

[KA8WTK]

Ah, the oil... I use good old 3-in-1 applied a drop at a time with a toothpick. Too much is not a good thing IMHO.

[WD8BIL]

One way to check if you have the "best" setting is to check the coil temp. There will be a lot of circulating currents and heat in the coil means loss!

[WA2ROC]

I agree that it would look great up on the wall, but not too close to the ceiling fan. 

Wouldn't want you to lose a hand in the "prop".

[Steve - K4HX]

You'll always want the setting that requires the least amount of coil/inductance and the largest amount of capacitance on the output (antenna side) cap.

One way to check if you have the "best" setting is to check the coil temp. There will be a lot of circulating currents and heat in the coil means loss!