Sanity check for wire stator

[CharlieM]

Hi all,
I’m planning another electrically segmented wire stator project to replace the great sounding but butt-ugly welding rod stators in my current beam splitter hybrids —this time using PVC insulated copper, stretched straight and glued to a nicer looking wooden ladder frame. The new panel will use .062 (1.5mm) d/s and will operate above 200Hz using the same 75:1 transformers and 2.7KV bias voltage.

In order to allow greater span between the ladder supports, I had planned to use .060” O.D. 18awg solid copper with .010” irradiated PVC insulation spaced 11 wires/inch for 40% open area. But I can’t find that particular wire anywhere, so my second choice is .052” O.D., 20awg with .010 irradiated PVC spaced at 13 wires/inch.

After reading some posts about the electrical effects of the PVC insulation shaping the static field and reading that Acoustat used tiny 24 awg wire with really thick insulation, I’m no longer sure that my assumptions are valid—so I need a sanity check before proceeding.

Assumptions:
- About 40% open area is optimal.
- Wire spacing can be less than but not greater than the d/s.
- For my application, .010 thick PVC should be adequate.
- More copper/less PVC per area equals higher efficiency. Hence; .
010 PVC should give greater output than .016 PVC (given same
open area).

Comments welcomed and appreciated.
Charlie

 

[CharlieM]

[No comments today] I have to decide whether to go with go with .010 irradiated PVC or standard .016 PVC insulation.

Once the insulation thickness is decided, that will determine the wire O.D. and thread pitch to set the wire spacing in the stretcher jig.

 

[WrineX]

solid wire or stranded ?

 

[CharlieM]

solid wire or stranded ?

solid wire would be my choice... my questions now are which gauge, insulation and spacing combination would would best fit the driving voltages.

 

[bolserst]

…After reading some posts about the electrical effects of the PVC insulation shaping the static field…

I don’t recall reading about this, can you post a link?

Assumptions:
- About 40% open area is optimal.
- Wire spacing can be less than but not greater than the d/s.

40% - 50% open area (with air space between wires < D/S) is a good compromise between efficiency and negative HF effects of added mass from the slugs of air in the slots between the wires, and cavity resonance. Once you start to get < 40% open area the HF cavity resonance will start to become more prominent, but not necessarily troubling. If you try to avoid the negative HF effects by making the stators more transparent(ie > 50% open area) you will start to lose sensitivity as air space between wires will start to exceed D/S.

- For my application, .010 thick PVC should be adequate.
- More copper/less PVC per area equals higher efficiency. Hence; .
010 PVC should give greater output than .016 PVC (given same
open area).

It is true that adding insulation thickness reduces sensitivity for a given stator-to-stator applied audio signal, but the losses are quite small.
For an example lets assume D/S = 1.5mm, and insulation = PVC with dielectric constant of 3 or 5.
The loss relative to an ideal un-insulated wire stator can be calculated from the resulting capacitive voltage divider:
- PVC insulation(k = 5, t = 0.010”) ===> -0.3dB
- PVC insulation(k = 5, t = 0.016”) ===> -0.5dB
- PVC insulation(k = 5, t = 0.032”) ===> -0.9dB

- PVC insulation(k = 3, t = 0.010”) ===> -0.5dB
- PVC insulation(k = 3, t = 0.016”) ===> -0.8dB
- PVC insulation(k = 3, t = 0.032”) ===> -1.4dB

In practice I think it comes down to what wire is readily available and works dimensionally for the D/S and %open area you are after. I personally use 0.016” insulation just because it is a little more resistant to failure form nicks/abrasions and adhesives during stretching process and assembly than 0.01”. My guess as to why Acoustat chose thicker insulation was to avoid failures from the solvents used during the process of attaching the wires to the frames.

***
I just noticed in your second post you were asking about choosing between 0.010" irradiated and 0.016" standard(non-irradiated?) PVC insulation. I made the assumption(bad) that all the insulation you were looking at was irradiated.

If those were the only two choices, I would definitely take the irradiated 0.010".

 

[CharlieM]

...I just noticed in your second post you were asking about choosing between 0.010" irradiated and 0.016" standard(non-irradiated?) PVC insulation. I made the assumption(bad) that all the insulation you were looking at was irradiated.

If those were the only two choices, I would definitely take the irradiated 0.010".

Thanks for your input, Sir!
I tend to agonize over every little detail, especially when I’m not confident about what I don’t know--- but I now feel more at ease with either choice.

BTW, my concern with insulation thickness was not its insulating value per-se impeding charge strength, but rather that thicker PVC necessitates increasing the [copper-to-copper] spacing between conductors to maintain 40% open area (hence; less conductor area carrying charge). With either choice (11 or 13 wires/inch, depending on insulation thickness) the copper-to-copper spacing won’t exceed the d/s so I should be good to go.

 

[bolserst]

…With either choice (11 or 13 wires/inch, depending on insulation thickness) the copper-to-copper spacing won’t exceed the d/s so I should be good to go.

Remember, compared to the air in the gap, PVC is a good conductor of DC and AC voltages. So, when building ESLs with PVC insulated wires the copper-to-copper spacing is not important, it is the PVC-to-PVC spacing that matters from an electrical stand point. (as well as acoustical)

Attached is a table of some different options for wire size, insulation thickness and wires/inch spacing.
The resulting %open area and the gap size between the wires are included.
Illustrations for Options “A “and “I” are shown to help clarify what the dimensions mean.
As long as your chosen D/S is > the gap between the wires, the sensitivity will be essentially the same for all options.

 

[CharlieM]

You're a life saver, Steve. Thank you so much. This exactly the kind of direction I needed!

 

[kazap]

Remember, compared to the air in the gap, PVC is a good conductor of DC and AC voltages. So, when building ESLs with PVC insulated wires the copper-to-copper spacing is not important, it is the PVC-to-PVC spacing that matters from an electrical stand point. (as well as acoustical).....

I wonder how magnet wire would work as stator wires?

The insulation on Beldon 8083 is a cross-linked, modified polyester that goes much higher then PVC in resisting electrical breakdown which might be useful?

The magnet wire is thin ( eg .0078in) allowing the wires to be spaced closer together for better field evenness while giving more open area.

It can be stretched to 15- 20% elongation with no insulation cracks.


But maybe thicker PVC acts like a cylinder under charge with the charge accumulating on the outer circumference of the PVC in a way that can create a descent field?

Any thoughts?

 

[geraldfryjr]

The info that Bolserst has shown in post 5 is excellent!!
This confirms my understanding of why Rodgers Sanders had expalined to me that if you had two coatings where one had a lower Dielectric Strength but had a higher Dielectric Constant (K) than the other, "Choose the one with the Highest K".

The higher the Constant will incraese the capcitance and lower the AC resistance thus increasing the efficiency as shown in Boserst's data.

Now in the past I have advised DIYer's to stray from paints that have Titanium Dioxide as a pigment due to its conductivity at high voltages and low volume resistance overall, compared to talc it is several magnitude of a difference.

But, it does have one very nice properity it has very High Dielectric Constan in the 80-100 range, This is some 20 to 50 times greater than the materials we have been using.

I am wondering if used in the right combination we could have aneven better coating.

I used such a primer that had large amounts of Titanium Dioxide as my base coat on bare metal but it was only one coat.

What if the coating of this materail was much thicker?

Since it is encapsulated in Acylic it does have some insulating factors.

Then to finish off the top coats with the Talc based primer and clear (like I have been doing) to give the needed HIgh Resistivity factor.

Could this actually yeild a bit more Efficiency?

Or maybe start off with the methods that I have been using to get my Dielectric Strength up and then throw on a heavy coat of Titanium Dioxide based coating to increase the capcitance.

Sorry if this seems a bit of OT. but it isn't really. 😉

jer 🙂

 

[bolserst]

I wonder how magnet wire would work as stator …The magnet wire is thin ( eg .0078in) allowing the wires to be spaced closer together for better field evenness while giving more open area…It can be stretched to 15 - 20% elongation with no insulation cracks.

Magnet wire seems like a perfect fit for ESL stator construction and depending on the insulation color can be visually appealing.
However, I ran into two problems when I tried it:

1) The insulation was too good of an insulator, so that the insulation coating the wires collected charge and the sound output would drop in level as a listening session went on. This was most noticeable during low humidity weather. In the summer it wasn’t much of a problem.
This issue has been discussed in a few threads:
http://www.diyaudio.com/forums/planars-exotics/58048-fresh-esl-research-2.html#post653417
http://www.diyaudio.com/forums/plan...-single-ended-esl-test-setup.html#post3118346
http://www.diyaudio.com/forums/planars-exotics/264972-measured-my-esl-attempt-3.html#post4140681

2) I ran into problems with insulation cracking on all of my stators. I used Essex Superior wire. Perhaps the Belden formulation would work better?
Some posts on the cracking issues here:
http://www.diyaudio.com/forums/planars-exotics/148309-wire-stator-design.html#post1899213
http://www.diyaudio.com/forums/planars-exotics/148309-wire-stator-design-4.html#post2248694

 

[kazap]

Magnet wire .....insulation coating the wires collected charge

Thanks. I wasnt aware of that problem so really appreciate your experience.

Mouser has four gauges of solid core copper PVC wire - 18/20/22/24

I wonder how thin the copper can be? Is 24 gauge solid core copper OK?

It seems plenty for current load and preventing audio frequency skin effects?