[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Terry's DRSSTC 6000 BPS testing



Original poster: Jimmy Hynes <jphynes@xxxxxxxxx>

Hi,

On 5/7/05, Tesla list <tesla@xxxxxxxxxx> wrote:
> Original poster: Terry Fritz <teslalist@xxxxxxxxxxxxxxxxxxxxxxx>
>
> Hi,
>
> >On 5/6/05, Tesla list Jimmy Hynes wrote:
> > > Original poster: Terry Fritz <teslalist@xxxxxxxxxxxxxxxxxxxxxxx>
> >
> >Can you fill in the rest of the details on this? like, all the ON/OFF
> >times, bps, energy/bang and all that? 20% compared to what? How were
> >you normalizing it?
>
> In the normal mode, the BPS could go from 40 to 300 BPS and the T1 time
> from 10uS to 500uS.
>
> In the burst mode, BPS could go from 1000 to 6000 BPS. The T1 time could
> go from 0uS to 300uS. There were 25 repetitions in each burst and they
> were spaced 1 second apart.
>
> The coil configuration draws roughly 200 primary amps and is described here:
>
> http://drsstc.com/~terrell/notes/DRSSTCspecifications.pdf
>
> A fairly recent model of the coil is here but does not have the burst mode
> implemented:
>
> http://drsstc.com/~terrell/notes/DRSSTC-22.sch
>
> The streamer was compared to the maximum length I could get in the normal > mode.


So you could get 20% more with 25 300uS bursts at 6000bps than 300bps at 500uS?

Where'd you come up with those numbers to start with? It's hard to
compare the numbers on a "Freau Number" basis, because you're only
doing a single burst pretty much. I guess the best way to compare it
would be spark length/peak IGBT temp. At 4ms, it might sink quite a
bit of heat into the case, but I don't know if it'd still be better to
go with one big bang, or possibly a few bigger bangs.

What IGBTs are you using? I tried to come up with a thermal time
constant for IRG4PC40W TO-247 IGBTs (ones I'm using) from the
'transient thermal impedance' chart. I came up with a little over a
mS, but it's tough to tell, because different points on the linear
portion of the curve still give different numbers. If the time
constant is ~1mS, then maybe the best is 1000 peak bps, with larger
bangs.

> At 05:25 AM 5/7/2005, Steve Conner wrote:
>
> >>The 6000 PBS burst mode seemed to produce roughly 20% longer arcs than
> >>the standard mode.
> >
> >At a cost of how much extra power consumption though? In other words what
> >was the "Freau Number" (spark length divided by square root power
> >consumption) for your coil in 6k bps pulse mode compared to say standard
> >200 bps?
> >
> >I would bet that it is less efficient (lower Freau number) at 6k than at
> >200, otherwise our theories on streamer growth are all wrong.
>
> The average time energy was far less since there is a 1 second pause
> between bursts.  It is difficult to know the actual burst energy since they
> are not discrete burst but rather a train of bursts joined together.  I am
> sure it can be figured out, but I have not worked it.
>

Try measuring the electrolytic voltage before and after.

>
> >>The DRSSTC protectors where the only real change and they seems to do the
> >>trick!
> >
> >What were those? Was that the extra inductors that you added or something?
>
> http://drsstc.com/~terrell/notes/DRSSTCprotec.pdf
>
> http://drsstc.com/~terrell/pictures/ArcFilter-01.JPG
>
> http://drsstc.com/~terrell/pictures/ArcFilter-02.JPG
>
>
> >>If anyone knows a super good way to compress video without killing sound
> >>and video quality I would like to know about it.
> >
> >MPEG format (.mpg file) seems to be best. DivX is good too but the free
> >version is adware. You can easily convert between formats, resolutions,
> >and frame rates with the free VirtualDub software. But you don't get
> >something for nothing, compression always hurts quality, you just have to
> >go for the biggest file that you think people will be bothered downloading.
>
> I am working on this...  In the burst mode, the video compressions do not
> capture the sparks well in many cases...
>

You could just do a 10 second clip instead of a 29 minute one. It'll
probably be enough for most people.
> Cheers,
>
>         Terry
>
> >Steve Conner
>
>