[TowerTalk] RG-213 Hi Pot ?

[jimlux]

On 5/10/19 10:15 AM, Steven Katz wrote:
> 10kW at 50 Ohms is only 707Vrms = 1kVpk.


The power rating is related to the current rating of the 13 AWG center 
conductor.

I = sqrt(P/R) = 14 amps for 10kW at 50 ohms.   Polyethylene has a low 
melting point and is a good thermal insulator.



What you're really interested in is the breakdown strength of the 
dielectric, and the field at the center conductor.  To a first order, 
you look at the radius of curvature of the center conductor, which is 
0.036" for AWG 13 or about 1mm. So, at 10kV, the field is about 10 
kV/mm, which sounds like a lot, and it would be, in air, where the 
breakdown is about 3 kV/mm.  But the breakdown for polyethylene (and 
most plastics) is about 20kV/mm (things with halogens, like PTFE, run 
about twice that).

This is for a solid dielectric - foam is MUCH lower - not only is the 
breakdown lower in the gas bubbles, but you've got field concentrations 
due to the non-uniform structure.

The breakdown in coax will also be lower where there's not a good shield 
- the shield helps keep the field stress uniform.

Real HV cable has a semiconductive layer around the center conductor, 
then the insulator - that helps manage the field distribution so you 
don't get concentrations, where corona can start.


The problem with HV cables is that they have microscopic defects, and 
the field concentration at a defect (when run close to the limit) causes 
a small discharge - that discharge breaks down the dielectric, creating 
a bigger defect, and it grows from there. Just like a crack in aluminum 
or plastic that grows on each stress cycle.  So you want to be a bit 
conservative using RF coax as HV cable - not abuse it, don't kink it, etc.

But RG-8 at 40-50 kV has been used for a long time.




> 
> -----Original Message-----
> From: TowerTalk <towertalk-bounces at contesting.com> On Behalf Of Mac
> Sent: Friday, May 10, 2019 10:05 AM
> To: towertalk at contesting.com
> Subject: Re: [TowerTalk] RG-213 Hi Pot ?
> 
> I'm wrong it seems [again] it is used in 10kw applications i have here .. but the specs sure dont support that from what i just reviewed..  im moving pretty fast as i am on a real project and might well be overlooking many parts of this ..  i will try later when i have more time ..  mac/mc  w5mc
> 
> On 5/10/2019 11:55 AM, Steven Katz wrote:
>> The mil spec rating for RG-213/U is 3700Vrms max; this is 5232Vpk.
>>
>> I wouldn't use it at 7.7kVdc.
>>
>> -----Original Message-----
>> From: TowerTalk <towertalk-bounces at contesting.com> On Behalf Of Jim
>> Thomson
>> Sent: Friday, May 10, 2019 9:52 AM
>> To: towertalk at contesting.com
>> Subject: [TowerTalk] RG-213 Hi Pot ?
>>
>> Does anybody know how much max V is on  RG-213 ?   Like what are the results if a hi-pot tester used..
>> and say cranked up to the point where  just a few ma occurs .     I have an application where I need a  HV
>> cable, less than 10 ft, that will  easily handle 5700 /  7700  vdc.   It will be hardwired at both ends, no HV connectors.
>> And with braid peeled way back from the center conductor, like  2-4 inches .
>> The plan was to also wrap some red electrical tape  every few inches  along its  entire  length, to readily identify it as a HV cable.
>>
>> Tnx....  JIm   VE7RF
>> _______________________________________________
>>
>>
>>
>> _______________________________________________
>> TowerTalk mailing list
>> TowerTalk at contesting.com
>> http://lists.contesting.com/mailman/listinfo/towertalk
>> _______________________________________________
>>
>>
>>
>> _______________________________________________
>> TowerTalk mailing list
>> TowerTalk at contesting.com
>> http://lists.contesting.com/mailman/listinfo/towertalk
>>
> _______________________________________________
> 
> 
> 
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
> _______________________________________________
> 
> 
> 
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>