Split Stator capacitors.

[jim_jobe]

Is there any fundamental difference between a twin- gang variable capacitor and a split stator type?
Jim

[Alf]

Possibly, on some, the moving veins are 180° apart and can rotate in opposite directions.

[AlanBeckett]

See http://www.mainlinegroup.co.uk/jacks....htm#catalogue
Alan

[jim_jobe]

Cheers for those. I always used to think the same; i.e. that they moved differentially like that, but I've never been absolutely sure as I've seen what looked to be exceptions.
Jim

[AlanBeckett]

Jim,
There is another variety of split stator capacitor where one set of moving vanes moves between two sets of fixed vanes which are opposite to each other. However, I can't find a decent picture
Alan

[jim_jobe]

Hi Alan,
I've got one of those. So, it seems that there are several types of split stator then. It's not always clear which type authors are talking about though,
Jim

[G8HQP Dave]

That would be a differential capacitor?

I have always been puzzled by split stator. I know at VHF you need to maintain balance so the butterfly capacitor is used, but at lower frequencies I can't see the difference between split stator and double-gang, except maybe that the split stator has no screen between the two stator sections. Is it just a case of high spacing, so high voltage, and no screen needed as the circuit is balanced? So you could use a double-gang for a low power transmitter, and you could use a split stator for a superhet but might get more oscillator radiation? You probably could not use a split stator for a TRF with an RF stage as there would be too much feedback.

[AlanBeckett]

I've always regarded the terms 'split stator' and 'butterfly', or even 'differential' as interchangeable. Multi-section (gang) variable capacitors are inherently 'split stator' in the sense that the individual stators are (electrically) separate, but the associated rotors are in line, and hence the capacitance of each section changes in the same 'direction'. With the former one section increases while the other decreases.
Is that reasonable, or is it more subtle than that?
Alan

[Skywave]

As I understand this topic, the essential differences go something like this:

Differential. The total capacitance of the capacitor remains constant, but as the shaft is rotated, for every incremental decrease in capacitance on one stator, the other stator undergoes an equal incremental increase.

Conventional two (or three gang *) split-stator capacitor.
As found in a conventional superhet. One section is always used for the oscillator - and this section may follow a particular law and have Cmax and Cmin values that are different from the 'signal' tuning sections. Nevertheless, as the shaft is rotated, the incremantal %-age change in capacitance of all the sections remains the same.

* I have seen - just once - a six-gang split-stator tuning capacitor!

'Transmitting' type split-stator.
On first analysis, it would appear to do the same job as a two-gang split-stator capacitor as above, but the essential difference is that the stators and rotors are set apart from one another not only by 180 degrees radially but also displaced longitudinally along the capacitor's axis. This is to maintain a true equality of capacitance in the two sections as the shaft is rotated and also to maintain equality in the strays from the stators to the rotors and from each stator and rotor to nearby metalwork which is usually at RF 'ground'. For a balanced ATU application, for example, such an arrangement is highly desirable - although I'm sure that the 'conventional' split-stator would do the job if the unwanted unbalances, including stray capacitances, could be lived with. In a nut-shell, it's a high-performance version of the conventional split-stator capacitor - and their prices usually reflect that! . Refer to the above Web page, "TX5 split-stator transmitting type" therein for a picture of one of these.

Butterfly.
Probably the most technically fascinating of them all. According to what I have read (F. E. Terman, for example, amongst others), this capacitor is more than a capacitor - or it's intended to be regarded and used that way. At first glance, it looks like a differential capacitor, but its intended use is for it to be so coupled into an inductance that as its capacity is varied, the inductance (albeit small) of the capacitor varies also - thus providing a much wider tuning range than could be achieved with a differential type of same capacitance - especially at VHF, where this type of cap. is commonly found in push-pull amplifier & oscillator circuits.

So that's my understanding, briefly, of the differences. It is not meant to be regarded as authoratitive! I am sure that there are a few errors in there somewhere - but I just thought I'd make my contribution to this fascinating topic & help it along its way.

Al. / Skywave.

[Ray Cooper]

Originally Posted by Skywave

...'Transmitting' type split-stator....but the essential difference is that the stators and rotors are set apart from one another not only by 180 degrees radially but also displaced longitudinally along the capacitor's axis. This is to maintain a true equality of capacitance in the two sections as the shaft is rotated and also to maintain equality in the strays from the stators to the rotors and from each stator and rotor to nearby metalwork which is usually at RF 'ground'....

A good summary. Just an addendum: what I've always regarded as the 'butterfly' type - widely used in high power TXs - has two sets of co-planar stator plates each of them covering about 90 degrees of the available rotation, with a rotor plate consisting of two 90 degree sectors placed 180 degrees apart. The 'rotational cycle' of this type is thus only 90 degrees, but it does offer advantages over the conventional construction.

Used in a push-pull arrangement - as these usually are - the rotor plate remains at earth potential throughout, and indeed doesn't need to be connected to earth physically. This has a big advantage, since it means that there doesn't need to be any sliding 'finger' connection to the rotor - always a problem when large amounts of RF are concerned, since they burn readily. When this happens, the RF tries to find its way out using the bearings, and usually does these a power of no good as well. Indeed, I've seen 'butterfly' types where the rotor was mounted on a glass rod spindle -totally no connection to the outside world, other than capacitive. The true 'butterfly' type has only two connections, both of them to the stators.

[jim_jobe]

What prompted me to start this thread is the attached circuit where VC1/2 is specified as a split stator 25 + 25pF capacitor but is obviously not a differential type. Why not just call it a twin gang? Usually it's obvious when a differential type is intended as the correct circuit symbol is used but this isn't always the case.
Jim

[Skywave]

Re: the 'butterfly' type capacitor.

There does appear to be some ambiguity surrounding this. Terman provides an excellent description and analysis ('Electronic & Radio Engineering', 4th. edition, page 500). However, that description refers to a butterfly capacitor in which the two stators are joined at their peripheries, each to each, by simply continuing the vanes of the stator round in a complete circle. This presents an inductance between the two 'lands' of the stators. Hence, the reference to varying the inductance of the capacitor as the shaft is rotated. In use, the whole capacitor actually becomes a variable capacitor and variable inductance: an additional coil is not required. Consequently, it seems more appropriate to refer to this 'variable capacitor' as a 'variable resonator'. On account of the variation of L and C over the range, Terman states that a frequency ratio of 5:1 is typically obtainable (at VHF). The resultant circuit affords a reasonably high Q, excellent mechanical stability and no sliding contacts.

Now this type of 'butterfly' I've never actually seen in reality - but I do own - (as I'm sure many of us have, or have at least seen) - the 'butterfly' capacitor in which the stators are not thus joined. Yes, the rotor is indeed 'butterfly' shaped, and yes, this capacitor is commonly used at VHF in balanced circuits. But the two types are not the same, mechanically or electrically.

Al.

[Skywave]

Originally Posted by jim_jobe

.
. . . . Why not just call it a 'twin-gang'? . . .
Jim

Which would be a reasonably sufficient description for all practical purposes; it's all a matter of using an established term. If you wanted to be pedantic, , you could argue that the term 'twin-gang' could refer to a capacitor in which the rotors and stators are each seperately insulated from each other and, therefore, the use of the term 'split-stator' - since this implies a common rotor - removes this ambiguity.

Al.

[G8HQP Dave]

Butterflies are not usually differential capacitors, although I suppose you could make a differential using the butterfly construction by having only half the rotor.

In Jim's circuit I would call VC1/2 a twin-gang. It is only a split stator in the sense that all twin-gangs have a split stator. I would have thought that the intersection screening of a twin-gang would be needed here to prevent too much capacitive coupling between the two tuned circuits, unless the designer intended this in which case only a particular split stator would have the correct capacitance between the stators.

[Skywave]

Originally Posted by G8HQP Dave

Butterflies are not usually differential capacitors. . .

Agreed. However, if you're referring to my earlier post, I did say "at first glance, they (butterfly caps.) appear to look like . . . ."

However, we are in agreement: 'nuff said!

Al.

[MichaelR]

Originally Posted by G8HQP Dave

Butterflies are not usually differential capacitors, although I suppose you could make a differential using the butterfly construction by having only half the rotor.

In Jim's circuit I would call VC1/2 a twin-gang. It is only a split stator in the sense that all twin-gangs have a split stator. I would have thought that the intersection screening of a twin-gang would be needed here to prevent too much capacitive coupling between the two tuned circuits, unless the designer intended this in which case only a particular split stator would have the correct capacitance between the stators.

Is Jims circuit essentially a colpitts local oscillator in which a split stator is required for simultaneous increase/decrease in the capacitors.A designated split stator is constructed such that it has less losses compared with a dual gang, again mentioned in some detail by another post

I have never come across differential

My understanding of butterfly is as SKYWAVE has pointed out that these were commonly used at VHF to get simultanoeus change in inductance and capacitance ( purposely designed to give such).

Mike

[Skywave]

The freq. changer in Jim's cct. is an interesting one. The screen-grid of the f.c. pentode is clearly acting as the anode of the oscillator and the tuning arrangements on g1 and g2 reflect that. Since there will be some inevitable capacitance coupling between L1 & L2, any stray capacitance coupling between the two capacitor stators will simply add to that. So, provided that all these stray capacitances are taken into account in the calculations for the L and C to achieve the required tuning ranges, it seems to matter little if the screening between the stators is not of a high order - assuming that there is a need for any screening at all.

Al.

[G8HQP Dave]

Agreed. However, if you're referring to my earlier post, I did say "at first glance, they (butterfly caps.) appear to look like . . . ."

Ooops - sorry, I read it too quickly!

I have never come across differential

I don't think I have ever seen one in the flesh, but I have seen them used in circuits. One use is in a VFO, to dial in variable amounts of temperature compensation - one stator has a positive temperature coefficient cap in series, the other stator has an equal value negative TC cap in series. Twiddling the differential cap does not change the frequency (at least to first order) but does vary the overall TC.

So, provided that all these stray capacitances are taken into account in the calculations for the L and C to achieve the required tuning ranges

The snag is that you would have to specify the exact cap, if your circuit was relying on a specific value of coupling. The stray coupling between the stators could be quite high compared with stray capacitance between the coils. It would be better to specify a twin-gang, then add a small top-coupling capacitor.

[jim_jobe]

I think Birkett's are selling differential air-spaced trimmers at 3 for £1 at the moment.
Jim

[Ray Cooper]

Just for the sake of completeness: attached is a photo of a differential trimmer, for the benefit of those who haven't come across these before.

As you can see: in this case, it has two sets of stator plates, but in this case they cover nearly 180 degrees of a circle. The rotor plate also covers about the same angle. Clearly, as the rotor is adjusted, the capacitance of one half of the device will rise, whereas that of the other half will fall by a similar amount. In the particular circuit in which this specimen was used (a bit of ex BBC gear) the trimmer was used to balance the overall capacitance of two halves of a bridge circuit.