Topics

Non Linear Vac Cap?!


Barry - 2E0IDE
 

Guys

I've just changed out an air cap for a vac cap and was surprised to find, unlike a very linear air cap, the vac cap is anything but linear.:-


My controller was really accurate with the air cap and a 100:1 gearbox. The vac cap has a 5:1 gearbox. At 40M the accuracy is OK but not good at all at 20M and above.

Has anyone else had similar results with a vac cap or can advide where I may be having a senior moment!


guenter
 

Hi,
some caps are really not nice (far from linear) in the lower region: 

Just check and compare: https://www.amstechnologies-webshop.com/media/pdf/8e/a1/a8/Vacuum-Variable-Capacitors-Jennings-Datasheets.pdf

73 guenter


Barry - 2E0IDE
 

Guenter

 

Thanks for the reply.

 

My Vac Cap is ex Russian – built 1991 and still in the box certified for 3KV but tested up to 5KV  and unfortunately is very non-linear.

 

I’m assuming the tuning code takes the next lower rang, next upper range and calculates accordingly. I’ll have to take a look and see if I’m correct. IF someone can point me to that section of code I’d appreciate it.

 

I hoped that adding 25KHz range steps at 20M would help but it did not .Perhaps I’ll have to add more.

 

73

 

Barry – 2E0IDE

 


Virus-free. www.avast.com


Immo, DL5KB
 

hi Barry, 
I tried to retrace your data as far as possible. My experience with vaccum-Caps is another, then your data show. I believe, that you have some slip in your actuation. I did a little calculation for frquency per step and saw several glitches:  i.e. from 14.125 to 14.150 are 8.704 steps, also from 14.2 to 14.225 you have documented 1.416 steps. There are some points more.
Is it possible, that the drive-shaft turns loose sometimes or you go with to high rpm, that the stepper looses steps (glitches)?
The vac-caps are normally linear from maximum C down to 5% of C. At the last part (min C) there start, depending on construction, non linearities.
If your cap is NOS, it is also possible (and normal), that the lubrication of srew thread has got old and stiff. You should renew the lubrication. 
And yes, the controller makes a linear interpolation of the steps between two stored values to get the value for a not stored frequency. So you have to take the more programmed steps, the nearer to the minimum of cap.
So take a look to your system.
Good luck!
73, Immo


Barry - 2E0IDE
 

Thanks for the feedback Immo.

 

I did consider slippage in the steeper. When I first replaced the air cap with a vac cap it had a non- geared stepper(i.e. 1.8 degrees) and I got very similar results. Thinking it was a torque problem I replaced the stepper with 5:1 geared stepper.

I can turn the vac cap by hand but there is quite a lot of resistance, so you could be right that there is still slippage,  even with the increase torque of the 5:1 gearbox.

 

There is a plastic cap on the drive end that looks as though it may be removeable. Perhaps this is where I can re-lubricate. Any thoughts?

 

Attached is a plot of the 20M band after pushing SWR tune and storing,  at each step – it’s linear across the band! But when I went back to 14.000 to check it, the SWR was NFG. So it does indeed look like slippage. Controller stepper settings are 60rpm, 4x, 4, so I don’t think it is a speed problem. Stepper backlash 300.

 

73

 

Barry – 2E0IDE

_,_._,_


Virus-free. www.avast.com


Immo, DL5KB
 

Barry,
your data as diagramm...
73
Immo


Barry - 2E0IDE
 

Immo

 

You’re a star! Thanks

 

Problem solved – 30 year old grease! I oiled the thrust bearing and screw thread with clock oil, then manually ran the cap full travel several times.  Re tuned all the range settings with the swr tune button and stored results. 14.0MHz to 14.1MHz was accurate to 24HZ which is plenty good enough for me.

 

73

 

Barry – 2E0IDE

 

 

 


Virus-free. www.avast.com


Immo, DL5KB
 

Barry,
your'e welcome. This experience is part of my worklife. 

By the way: beautiful work, your tuning-box.
Perhaps a little remark to this solution. The connection to the head of the capacitor elongates the loop on this side. So the mechanical middle of the loop is shifted half the length of this connection to this side. It may (I say it may!) change the reachable best SWR, because the coupling loop now is not at the point of zero of the loop. Using a gamma-match, you should search the new zero on the loop, with a coax coupling-loop it may have no such importance.
With an earlier built, I used a gamma-match and had to shift it abt. 3 inches for good result. The picture shows my first solution with a vac-cap on the bench, just with an ungeared Nema 19. My loop today has a 5:1 geared stepper and a coax coupling-loop.
73, Immo


Barry - 2E0IDE
 

Immo

 

Thanks for the comments and advice. This is my first attempt at building a Loop. I used a modified “Work Mate “ to bend the copper tube as per pic(an idea from  John G6KJK) with heat shrink instead of paint. All copper fittings were hand-made, and all plastic mounts and loop fitting were home designed and 3D printed.

 

I did wonder about the copper band on the drive end and will take your advice and re-design.

 

I’m using a coax Faraday Loop coupler that is a slightly squashed circle (squashed even more than the pic - It does improve SWR). The coupler is designed to be easily removable so that I can experiment with gammas etc.

 

After the lubricant fix on the vac cap, the SWR on 20M is negligible but rises slightly for the higher bands.

 

I’ve had good contacts into Eastern Europe on my meagre 50 watts allowance.

 

Overall, a very enjoyable project and the controller is to be commended. Arduino based rotator next.

 

Once again thanks for all the help and advice.

 

73

 

Barry - 2E0IDE

 

 

 

From: loopController@groups.io <loopController@groups.io> On Behalf Of Immo, DL5KB
Sent: 07 March 2021 15:03
To: loopController@groups.io
Subject: Re: [loopController] Non Linear Vac Cap?!

 

Barry,
your'e welcome. This experience is part of my worklife. 

By the way: beautiful work, your tuning-box.
Perhaps a little remark to this solution. The connection to the head of the capacitor elongates the loop on this side. So the mechanical middle of the loop is shifted half the length of this connection to this side. It may (I say it may!) change the reachable best SWR, because the coupling loop now is not at the point of zero of the loop. Using a gamma-match, you should search the new zero on the loop, with a coax coupling-loop it may have no such importance.
With an earlier built, I used a gamma-match and had to shift it abt. 3 inches for good result. The picture shows my first solution with a vac-cap on the bench, just with an ungeared Nema 19. My loop today has a 5:1 geared stepper and a coax coupling-loop.
73, Immo


Virus-free. www.avast.com


K4MJA
 

Interesting! I have the same or very similat cap and it is hard to turn. I will be try to lube it tonight!