[Ian Bell[_2_]]

flipper wrote:
On Thu, 18 Dec 2008 11:37:33 +0000, Ian Bell
wrote:

flipper wrote:
On Wed, 17 Dec 2008 22:08:06 +0000, Ian Bell
wrote:

Raymond Koonce wrote:
Ian Bell wrote:
Raymond Koonce wrote:
Hi RATs,

I finally got some photos posted from ETF 08. Take a look at
http://www.timebanditaudio.com/ETF08/etf2008.html

Also visit the ETF website for additional links to photos
http://www.triodefestival.net/index....tf-2008-photos. This
was a great meeting.

Regards,

Raymond
I notice on the Agenda that there was a session by Aernoud Dekker on
LTSpice and also on using curvetracer for the creation of tube models.
Do you know if it is possible to obtain copies of this presentation??

Cheers

Ian
Hi Ian,

His presentation is not listed on the ETF site, but I'll drop him an
email and see if he can share it. It was very interesting, but I'm
afraid I didn't listen closely enough.
Thanks for that. Fingers crossed.

I'm still poking around trying
to figure out how to get my triode model worked into the schematic.
Curvetracer looks very interesting, in that you can take a scan of any
tube's curves and digitize it for use in LTSpice.

Is it a commercial product? Does it have a web site? A Google search for
curvetracer brings up loads of hits but none seem to fit your description..
I think they're talking about Curvecaptor, which works by 'curve
tracing' the triode plate curves.

http://sourceforge.net/projects/curvecaptor/

With Windows you also have to install ActiveTcl. I don't remember
exactly where I got the one I'm using but I think you can get it here.

http://www.activestate.com/Products/...cl/index.mhtml

You run Curvecaptor via Curvecaptor.tcl

You make a pic of the tube curves, load it into Curvecaptor, and then,
essentially, 'trace' the curves dot by dot. The program records the
X-Y coordinates of each as you do . You're then offered a choice of
models (well, it defaults to one but there's a drop down menu for
more), you pick which model you want to use and it shows you a plot
generated from the model picked so you can see how well it does. You
can change the model type and repeat the plot process as often as you
like in order to compared and pick which model you think does the best
job. It'll then generate the two versions of spice code and you
cut/paste the one for your spice program into an appropriate text
file.

Note that it just makes curve equations and there is no grid current
modeling. Oh, manually add the capacitances using an existing model as
a guide for the format.

The pic needs to be a GIF and I forget what the 'maximum' dimensions
are but looking at the ones I made it's about 700x550. Smaller reduces
the resolution but bigger doesn't 'fit' inside the window allowed. I
think I generally set width to 700 and the height, due to the curve
graph itself, just came out whatever it was. Oh, and you do need the
axis numbers because that's going to be one of the 'traces' (so it can
correct for imperfect alignment and, of course, it needs to know what
values the 'dots' are being plotted against).

Tedious but relatively simple once you've made it through one.

Cheers

Ian
Interesting. I downloaded it and ran it on my Linux box - it worked OK
but it had no documentation and I could not work out what sort of file
it wanted. Does the Windows version have more documentation?

Nope.

See my other post. What it's asking for when first opened is the tube
curves GIF (at least on windows) file. I mean, a 'picture' of the
curves, in GIF format, that you'll then use to 'graphically' pick data
points from.

Yes, saw your other post. Useful little tutorial that.


You can save data at each step and then 'resume' from there. Like, if
you've done the curve tracing and saved the .DAT file you can kill the
opening GIF file dialogue, go to the data step, and load the .DAT
file. I mention that because you mentioned wanting to load in
someone's data so if you look at the .DAT file format you might be
able to come up with a way of just transferring the data into one and
go from there.


Yes, the .DAT file looks promising; I'll check out its format. I also
emailed the author so maybe I'll get some insight from that source too.

Cheers

ian



Cheers

Ian
Best regards,

Raymond