Magnetic properties of silicon iron laminations
Winfield Hill
John Woodgate, <j...@jmwa.demon.co.uk> said...
>
> I have measured a small (5/8 inch square stack of 0.020 inch lams)
> nominally 1:1 transformer (a fairly vintage RS Components 'Midget
> speaker isolating transformer') with applied voltage proportional
> to frequency, thus maintaining (nearly) constant induction ('flux
> density'). I have used two levels of excitation, 3 V at 1 kHz and
> 0.75 V at 1 kHz - the latter might show some 'bottom bend' effects
> due to low induction. [ snip ]
>
> From the measurements, the impedance magnitude and phase angle were
> calculated, leading to the series equivalent XS and RS. The d.c.
> resistance (0.36 ohm) and the 0.1 ohm were subtracted from RS, and the
> resultant XS and Rs converted to parallel equivalents Rp,and Xp. This
> is in line with the normal presentation of iron loss as a resistance
> in parallel with the magnetizing inductance Lp. Rp is composed of
> hysteresis loss and eddy current loss. It 'can be shown' that, at
> frequencies where eddy-current loss is significant, both Rp and Xp
> rise as the _square root_ of frequency.
Very interesting data, John. Yes, that's possible, with rising Rp
due to the eddy-current loss and with a drop in expected Xp due to a
drop in permeability from eddy-current shielding in the core.
> ... The measurements at low frequencies are not very repeatable:
> I put this down partly to the low signal levels and partly to residual
> magnetization effects.
>
> It can be seen that over a considerable frequency range, Rp and Xp are
> approximately proportional to frequency, while at higher frequencies the
> gradients fall to near the orthodox 'square root' value. Up to about 2
> kHz the permeability is roughly constant, and the main effect on it
> seems to be rising eddy current loss: when the loss resistance and the
> reactance are comparable in value, the permeability starts to rise less
> than proportionally.
Very nice John, showing a 2x higher Xp and thus permeability, for a 4x
higher magnetization flux density, B, assuming you didn't slip a factor
of 2 someplace! Is this a SQRT B relationship or is that coincidence?
Hmmm, another prominant mechanism for rising Rp is that old bugaboo
proximity-effect wire losses, which also goes as the square root of
frequency (for d < 5 delta). I notice your Rp losses are independent
of amplitude, and start a distinct SQRT f slope above 250Hz, while the
Xp slope remains a nice inductor, proportional to frequency, and doesn't
droop to SQRT f until 2kHz or so... Could a dominant Rp mechanism in
your measurements be proximity effect (which can get very serious at
even low frequencies, for coils with more than 5 to 15 layers), setting
in for the wire at low frequencies, well before the core laminations are
too thick to keep eddy-current shielding at bay?
--
Winfield Hill hi...@rowland.org _/_/_/ _/_/_/_/
The Rowland Institute for Science _/ _/ _/_/ _/
Cambridge, MA USA 02142-1297 _/_/_/_/ _/ _/ _/_/_/
_/ _/ _/ _/ _/
http://www.artofelectronics.com/ _/ _/ _/_/ _/_/_/_/
Winfield Hill
Chuck Parsons, <Ch...@CatenaryScientific.com> said...
>
> Winfield Hill wrote:
>>
>> Very nice John, showing a 2x higher Xp and thus permeability, for a 4x
>
> Very nice stuff John. I am sure it is just a coincedence. I attach
> a 11K gif file I made from data from the CRC on permeability.
> You can see if the transformer is less than about 6000 guass
> at 3volts, the factor of 2 is very plausable.
Too bad that graph starts at 2000, aren't we especially interested
here in the region from -2000 to 0 to +2000 ....
-- Win
Chuck Parsons
Sorry, that's all the data in the book, still I think half a loaf
is much better than none.
The only other info, and perhaps it has some interest is
"Initial Permeability of High Purity Iron for Various Temperatures"
Temp C Permeability
0 | 920
200 | 1040
400 | 1440
600 | 2550
700 | 3900
770 |12580
So it looks like the iron isn't going to be a problem in a high temp
transformer. And hopefully the above table gives us a guess at the Y
intercept. It doesn't say what they used for "Initial" and they
don't give data for silicon steel :-(
Wait! Another table gives permeability at 20 Gauss.
P ( 20 Gauss) P (Max) Saturation(Gauss)
4% Silicon Iron 500 7,000 19,700
(Grain Oriented) 1500 30,000 20,000
So there is some where between a 13.5-1 to 4.5-1 ratio of
permeability.
That is going to make the leakage inductance and the magnetizing
inductance vary a lot if there is any dynamic range in the excitation.
Second, based on the graph and the CRC value of saturation at 2Tesla
I see that there is saturation, and then there is SATURATION.
Hmm, at some point people where talking about using lower excitation
voltages to measure stuff. Sounds like a marketing trick to me. If
a FeSi transformer is reaching 1.4 Tesla. A 20% decrease more than
doubles the permeability. That has to make a big difference in the
leakage inductance.
John Woodgate
In article <34BABD0A...@CatenaryScientific.com>, Chuck Parsons
<Ch...@CatenaryScientific.com> writes
>Winfield Hill wrote:
>>
>> Very nice John, showing a 2x higher Xp and thus permeability, for a 4x
>> of 2 someplace! Is this a SQRT B relationship or is that coincidence?
>>
>
>You can see if the transformer is less than about 6000 guass
>know! Iron is very non-linear!
>
>Chuck
>[ A MIME image / gif part was included here. ]
>
Unfortunately, I can't UU-decode your .GIF. I don't know why, maybe it
was corrupted in transmission.
I don't know the induction value, but I can squeeze a few extra turns
on, so I can calculate it.
OK, I did that. I added 21 turns and, BTW, noted that it's a 9/16 inch
(or 14 mm) square stack, not 5/8 inch as I said before. Applying 3 V to
the input winding I get 1.23 V across 21 turns. Using the well-known
formula V = BA[omega]N, rearranged, B = 48 mT or 480 Gs. That checks
with the figures in my 'little black book' for 240 V 50 Hz primary
windings on that size stack.
Thinking about it, it couldn't be near 6 kGs at 1 kHz, because this is a
(admittedly small) loudspeaker transformer, and should at least handle 3
V input (3 W in 3 ohms) at 150 Hz or so, which would give 40 kGs! 48 mT
at 1 kHz leads to around 1 T at 50 Hz, which I could believe.
--
Regards, John Woodgate, Phone +44 (0)1268 747839 Fax +44 (0)1268 777124.
OOO - Own Opinions Only. You can fool all of the people some of the time, but
you can't please some of the people any of the time.
John Woodgate
In article <69epp9$p...@fridge.shore.net>, Winfield Hill
<hi...@rowland.org> writes
>Chuck Parsons, <Ch...@CatenaryScientific.com> said...
>>
>> Winfield Hill wrote:
>>>
>>> Very nice John, showing a 2x higher Xp and thus permeability, for a 4x
>>
>> Very nice stuff John. I am sure it is just a coincedence. I attach
>> You can see if the transformer is less than about 6000 guass
>> at 3volts, the factor of 2 is very plausable.
>
> here in the region from -2000 to 0 to +2000 ....
>
>
Murphy Rules! I can't UU-decode the .GIF! (8-(
James Meyer
On Mon, 12 Jan 1998 20:02:02 -0500, Chuck Parsons
<Ch...@CatenaryScientific.com> wrote:
> Very nice stuff John. I am sure it is just a coincedence. I attach
>a 11K gif file I made from data from the CRC on permiability.
You could have saved about 968 bytes by encoding the .gif with
only 2 colors instead of the 256 you used. 8-)
Jim
Chuck Parsons
I tried to keep it small. IMSI Graphics Converter Gold,
at least as far as I can tell won't output CGM files, though
it reads them. I tried converting the image to 1Bit then saving
it as GIF, same result it used 8bit.
Fortunately, this file started out as black and white. The color,
compression option in IMSI aren't very good for turning something into
black & white. What I would like for schematics is background white,
everything else black, but it wants to halftone or something else
which isn't appropiate for line drawings.
Any suggestions on a program?
Especially one that doesn't cost a lot...
Mike
Chuck Parsons wrote:
[...]
> Any suggestions on a program?
>
> Especially one that doesn't cost a lot...
You might try Paint Shop Pro at http://www.jasc.com
Best Regards,
Mike
CEO, Analog & Digital Design
Automated Production Test
http://www.csolve.net/~add/home.htm
Hosting Jonathan Ramsey's Pascal TCP/IP for DOS:
http://www.csolve.net/~add/zips/tcp.htm
Winfield Hill
Chuck Parsons at Ch...@CatenaryScientific.com says...> I downloaded it and I'm giving it a try. Seems O.K.
> though it doesn't do CGM. It has something
> called PNG "Portable Network Graphics"
Hmm, I'm surprised I don't have a late version of HiJaak
or PSP on my computer at work, will try it at home...
--
Winfield Hill hi...@rowland.org
Rowland Institute for Science
Cambridge, MA 02142
Mike
Chuck Parsons wrote:
> Mike wrote:
[...]
> > You might try Paint Shop Pro at http://www.jasc.com
> I downloaded it and I'm giving it a try. Seems O.K.
> though it doesn't do CGM. It has something
> called PNG "Portable Network Graphics" I tried it
> and my graph went to 5K with no loss in clarity.
> At only 5K I'm going to post it, in hopes that others
> can also read it. Perhaps it is as good as CGM?
Isn't PNG great? It's free, which means companies don't have to pay
royalties for GIF. The decode is no more difficult than GIF, JPEG, or the
rest of the compressed formats.
A large number of DOS and Windows graphics viewers can read it.
But none of the browsers I tried can :(
Mike
Chuck Parsons wrote:
[...]
> I downloaded it and I'm giving it a try. Seems O.K.
> though it doesn't do CGM. It has something
> called PNG "Portable Network Graphics" I tried it
> and my graph went to 5K with no loss in clarity.
> At only 5K I'm going to post it, in hopes that others
> can also read it. Perhaps it is as good as CGM?
Chuck,
Just wanted to let you know the attachment "Perm.png" came through fine.
It shows up great on all my viewers.
John Woodgate
In article <34BB968D...@CatenaryScientific.com>, Chuck Parsons
<Ch...@CatenaryScientific.com> writes
>James Meyer wrote:
>>
>> On Mon, 12 Jan 1998 20:02:02 -0500, Chuck Parsons
>> <Ch...@CatenaryScientific.com> wrote:
>>
>> > Very nice stuff John. I am sure it is just a coincedence. I attach
>> >a 11K gif file I made from data from the CRC on permiability.
>>
>> You could have saved about 968 bytes by encoding the .gif with
>> only 2 colors instead of the 256 you used. 8-)
>>
>> Jim
>
>
> I tried to keep it small. IMSI Graphics Converter Gold,
>at least as far as I can tell won't output CGM files, though
>it reads them. I tried converting the image to 1Bit then saving
>it as GIF, same result it used 8bit.
According to page 3-7 of the manual, it will. I agree that it doesn;t
always do what you tell it to.
Chuck Parsons
John Woodgate wrote:
>
> In article <34BB968D...@CatenaryScientific.com>, Chuck Parsons
> <Ch...@CatenaryScientific.com> writes
> >at least as far as I can tell won't output CGM files, though
> >it reads them. I tried converting the image to 1Bit then saving
> >it as GIF, same result it used 8bit.
>
> According to page 3-7 of the manual, it will. I agree that it doesn;t
> always do what you tell it to.
> >
Apparently we have different versions. I have 3.0. The is no
page 3-7 on the manual chapter three ends on 3-6. I checked
all the Export output file options, no CGM. Did you spend extra
for the internet upgrade?
Tony Williams
In article <34BB81EA...@CatenaryScientific.com>, Chuck Parsons
<URL:mailto:Ch...@CatenaryScientific.com> wrote:
[snip]
> Hmm, at some point people where talking about using lower excitation
> voltages to measure stuff. Sounds like a marketing trick to me.
'twas moi, and I had no sinister marketing motive. It comes from much
experience of measuring certain magnetic properties, (Imag, B-H loss
and eddy-loss, permeability) and devising a methodology whereby the
results are reasonably consistent between three sites, all using
different equipment, different search-turns, and stimulus source-Z.
This thread is also about doing such basic measurements and then applying
the results to forward prediction of what the transformer will do when
used in a power supply.
My suggestion was simply to do all upfront theoretical measurements at
about 80% of full stimulus so as to avoid getting caught in any high
I-mag and/or 3rd harmonic current ambush. In order to compare theory
with practice in the final supply this also means that the supply
stimulus has to be at 80% of fsd.
Ok, that's not real life, but this thread is about accurately equating
the basic measurements to the final performance. It therefore seems
prudent to evade (or factor in) any (distorted) primary shunt-current.
> If a FeSi transformer is reaching 1.4 Tesla. A 20% decrease more than
> doubles the permeability. That has to make a big difference in the
> leakage inductance.
The books say that most of the leakage inductance is in air, so any
variation in the permeability of the core-material should have only
second-order effects. But maybe......?
--
Tony Williams, Ledbury, UK.
Winfield Hill
Chuck Parsons, <Ch...@CatenaryScientific.com> said...
> though it doesn't do CGM. It has something
> called PNG "Portable Network Graphics" I tried it
> and my graph went to 5K with no loss in clarity.
> At only 5K I'm going to post it, in hopes that others
> can also read it. Perhaps it is as good as CGM?
>
Hey, this worked great. The .PNG file was only 4,394 bytes
long, while rewritten as a .GIF file by HiJaak (which looked
identical, btw) the length was 10,886 bytes, or 2.5x longer.
Saved as a .JPG it took 164,084 bytes, or 37x longer, and
was damaged a bit.
So the .PNG is indeed pretty slick!! Now, if we can just
get all our readers out there up to speed!
Chuck Parsons
Mike wrote:
>
> Chuck Parsons wrote:
>
> > Mike wrote:
>
> [...]
>
>
> > I downloaded it and I'm giving it a try. Seems O.K.
> > though it doesn't do CGM. It has something
> > called PNG "Portable Network Graphics" I tried it
> > and my graph went to 5K with no loss in clarity.
> > At only 5K I'm going to post it, in hopes that others
> > can also read it. Perhaps it is as good as CGM?
>
> royalties for GIF. The decode is no more difficult than GIF, JPEG, or the
> rest of the compressed formats.
>
Chuck Parsons
Tony Williams wrote:
>
> In article <34BB81EA...@CatenaryScientific.com>, Chuck Parsons
> <URL:mailto:Ch...@CatenaryScientific.com> wrote:
> [snip]
> > Hmm, at some point people where talking about using lower excitation
> > voltages to measure stuff. Sounds like a marketing trick to me.
>
> 'twas moi, and I had no sinister marketing motive. It comes from
Tony, no slur intended! Just a joking way to say that leakage
inductance should get lower as the permeability goes up.
> My suggestion was simply to do all upfront theoretical measurements at
> about 80% of full stimulus so as to avoid getting caught in any high
> I-mag and/or 3rd harmonic current ambush. In order to compare theory
> with practice in the final supply this also means that the supply
> stimulus has to be at 80% of fsd.
Well until this thread I had no idea just how yucky ( to use the
technical term) transformer life is. Oh sure there is that constant
permeability region between 6.9 and 7.1 Tesla! ;-). Personally,
seeing the curves, as well as the variations for grain aligned versus
normal etc, I could well see the value in taking measurements, even at
points that were far from the operating conditions, but were less
prone to error.
After all the accurate spice model isn't the only criteria, much of
life is comparative/uniformity driven. If you want to look for QC
problems, or if you want to compare coupling, stable relative
measurements are better. And for that matter if you are looking for
problems with your spice model, stable measurements are absolutely
key.
One question I have, is how come some of the Audio fanatics, don't
run a DC current through (massively oversized) transformers, to stay
in a region of constant permeability? Hmm, the measurements I posted
were not small signal ( I think, CRC didn't say) maybe it looks quite
different if you do that?!
>
> Ok, that's not real life, but this thread is about accurately equating
> the basic measurements to the final performance. It therefore seems
> prudent to evade (or factor in) any (distorted) primary shunt-current.
>
> > If a FeSi transformer is reaching 1.4 Tesla. A 20% decrease more than
> > doubles the permeability. That has to make a big difference in the
> > leakage inductance.
>
> The books say that most of the leakage inductance is in air, so any
> variation in the permeability of the core-material should have only
> second-order effects. But maybe......?
>
I think that the relative permeability of the core and air have to
be very important in determining the leakage inductance, as well as
the magnetizing inductance. For leakage inductance, to my mind the
flux is making a choice air/core, higher permeability core will leave
less in the air.
BTW even though I can find old homework problems involving air/ high
permeability B-field energy calculations I didn't appreciate just how
much of the energy can be stored in the gap versus the iron. Certainly,
it should of jumped out at me because of the strong snap at the
end of a solenoid motion as the gap gets small.
Having reviewed that a bit on the theoretical side it is
interesting. Though, "all the energy is stored in the gap" is far to
strong a statement. And for big gaps, the tendency of the flux lines
to bulge/swell in the gap works against that argument as well.
Since magnetic energy is proportional to Integral(BH) the
permeability of the core works to reduce the integral since H is
low. But the concentration of flux by the core works to increase
it. Certainly, for constant current and winding diameter, there is
more energy storage with a core, than without ( 100 % gap.). If you
add a small dot of ferrite to a air core transformer, it will work to
increase energy storage in the vicinity of the ferrite, as the flux
lines are compressed.
Chuck
Kendall Castor-Perry
In article <34BCD13E...@CatenaryScientific.com>, Chuck Parsons
<Ch...@CatenaryScientific.com> writes
[big snip]
>
> One question I have, is how come some of the Audio fanatics, don't
>run a DC current through (massively oversized) transformers, to stay
>in a region of constant permeability? Hmm, the measurements I posted
>were not small signal ( I think, CRC didn't say) maybe it looks quite
>different if you do that?!
>
I don't count myself as an audio fanatic, but I can indeed report that
the current trend is for using single-ended tube amplifiers which run
the standing current of the output tube (usually a 300B, whatever that
is...) through the primary of the (massively oversized) output
transformer.
You simply would *not* believe what some people are prepared to spend on
the most esoteric of these amplifiers, but the fact that the windings on
some of the transformers are made entirely from pure silver might give
you an idea (-8
>>
[snip rest of the good stuff]
Kendall Castor-Perry
If there's a NOSPAM in the reply-to, remove it...
my employer pays for my net connection but I speak for myself.
Tony Williams
In article <34BCD13E...@CatenaryScientific.com>, Chuck Parsons
>
> Tony Williams wrote:
> >
> > In article <34BB81EA...@CatenaryScientific.com>, Chuck Parsons
> > <URL:mailto:Ch...@CatenaryScientific.com> wrote:
> > 'twas moi, and I had no sinister marketing motive. It comes from
> Tony, no slur intended! Just a joking way to say that leakage
> inductance should get lower as the permeability goes up.
No inferred slur seen.
> One question I have, is how come some of the Audio fanatics, don't
> run a DC current through (massively oversized) transformers, to stay
> in a region of constant permeability? Hmm, the measurements I posted
> were not small signal ( I think, CRC didn't say) maybe it looks quite
> different if you do that?!
I think that any offset area in the B-H loop would have far more
asymetry (aka distortion) than running it normally. Don't forget also
that the range of signal amplitudes means that you have to operate
over a range of B-H loop sizes, during small signals that may mean
running around a minor B-H loop that is quite nonlinear.
However, I have a vagrant thought crawling around in the hindbrain,
about the reason why magnetic recording heads used to run with a
large-amplitude high-frequency bias. Ah no, not the same thing,
that was to get rid of magnetic hysteresis, similiar to crossover
distortion. Memories of the Barkhausen Effect surfacing now.
> > The books say that most of the leakage inductance is in air, so any
> > variation in the permeability of the core-material should have only
> > second-order effects. But maybe......?
> >
>
> I think that the relative permeability of the core and air have to
> be very important in determining the leakage inductance, as well as
> the magnetizing inductance. For leakage inductance, to my mind the
> flux is making a choice air/core, higher permeability core will leave
> less in the air.
>
> BTW even though I can find old homework problems involving air/ high
> permeability B-field energy calculations I didn't appreciate just how
> much of the energy can be stored in the gap versus the iron. Certainly,
> it should of jumped out at me because of the strong snap at the
> end of a solenoid motion as the gap gets small.
>
> Having reviewed that a bit on the theoretical side it is
> interesting. Though, "all the energy is stored in the gap" is far to
> strong a statement. And for big gaps, the tendency of the flux lines
> to bulge/swell in the gap works against that argument as well.
If you take an ungapped inductor and cycle it through a moderately
large B-H loop, measure Imag, then separate-out and discard the
component due to watts-loss, the current left is the inductive
current and indicates the storage capability of the core itself.
It is not very large. 'All the energy stored in the air gap'
is a reasonably good working assumption on a gapped inductor.
John Woodgate
In article <34BC0312...@CatenaryScientific.com>, Chuck Parsons
<Ch...@CatenaryScientific.com> writes
>John Woodgate wrote:
>>
>> In article <34BB968D...@CatenaryScientific.com>, Chuck Parsons
>> <Ch...@CatenaryScientific.com> writes
>> > I tried to keep it small. IMSI Graphics Converter Gold,
>> >at least as far as I can tell won't output CGM files, though
>> >it reads them. I tried converting the image to 1Bit then saving
>> >it as GIF, same result it used 8bit.
>>
>> According to page 3-7 of the manual, it will. I agree that it doesn;t
>> always do what you tell it to.
>> >
>
> Apparently we have different versions. I have 3.0. The is no
>page 3-7 on the manual chapter three ends on 3-6. I checked
>all the Export output file options, no CGM. Did you spend extra
>for the internet upgrade?
No. I have v.2.1! Looks like someone did a downgrade!
John Woodgate
In article <34BCD13E...@CatenaryScientific.com>, Chuck Parsons
<Ch...@CatenaryScientific.com> writes
>
> One question I have, is how come some of the Audio fanatics, don't
>run a DC current through (massively oversized) transformers, to stay
>in a region of constant permeability? Hmm, the measurements I posted
>were not small signal ( I think, CRC didn't say) maybe it looks quite
>different if you do that?!
If you apply d.c and a smaller a.c. simultaneously, a small hysteresis
loop opens around the d.c. working point. That inevitably means that the
permeability varies over the a.c. cycle, otherwise - no loop!
Daniel Haude
On 14 Jan 1998 13:06:13 GMT,
Winfield Hill <hi...@rowland.org> wrote:
> So the .PNG is indeed pretty slick!! Now, if we can just
> get all our readers out there up to speed!
I'm using the SLRN newsreader on a Linux machine. It doesn't handle
attachments, so I'd have to save the message to a file, and then run some
decoder program to extract the PNG file. What program could I use for
that? Is there a PNG viewer that runs without X?
(My computer is a 386 w/ 8MB Ram. No X-Windows, no Netscape. Maybe that
explains my fondness of ASCII art ;-)
**Daniel
Winfield Hill
Daniel Haude, <p17...@public.uni-hamburg.de> said...
I suspect the .PNG standard was largely developed by folks running
UNIX etc. Many were internet programmers who were unhappy with UNISYS'
belated pursuit of a patent underlying the .GIF format, the sellout deal
CompuServe made with the UNISYS lawyers, and wanted a free open standard
with improved performance and capability anyway. Possibly some of the
.PNG programs they wrote don't require X-windows. How hard have you
looked? Hmmm, don't you need to step up a level anyway?
Dan, does it take more than 8M to run X-Windows? I'm pretty sure I ran
X-Windows (under Windows 3.1 !) while I still had a 386. OK, that was a
long time ago, maybe it was a 486, but hey, used 486 motherboards are all
over the place gathering dust, and another 8M of ram shouldn't be hard
to obtain. Especially old-style ram SIMMs. Oh, I see a problem, many of
those were only 1 or 2M, and you'd need 4x 4M. Another suggestion, new
motherboards with 120MHz Pentiums, etc. are pretty cheap right now.
That would make Linux really scream.
Daniel Haude
On 15 Jan 1998 13:06:47 GMT,
Winfield Hill <hi...@rowland.org> wrote:
> Possibly some of the .PNG programs they wrote don't require X-windows.
> How hard have you looked?
Not at all so far because I haven't come across a .PNG file yet. Really,
there is no serious trouble. I just didn't want to give up my defense of
ASCII art that quickly ;-)
> Hmmm, don't you need to step up a level anyway?
It looks as if you have a point there, but the situation is a little
different. I said I owned only a '386 8MB Linux PC at home, which is true.
This doesn't mean that I don't have two biggish Pentium machines right
next to it under my desk; I just don't own them (they belong to my roomate
who is currently doing a job in San Francisco).
> Dan, does it take more than 8M to run X-Windows? I'm pretty sure I ran
> X-Windows (under Windows 3.1 !) while I still had a 386.
Admit it: You didn't "run" it, I'd say you rather "crept" it. But anyway,
I started in on this thread more or less to take the case for ASCII art.
For this I presented myself (quite successfully!) as a poor computer stone
age fellow -- without actually lying to you. It's just that I didn't tell
you right away that I don't read news at home anyway but at the University
on an SGI Octane workstation. Feel free to call it my problem when my
favorite NR only runs in a Unix shell window...
OK, everybody flame me ;-)
**Daniel
Jan Panteltje
> Dan, does it take more than 8M to run X-Windows? I'm pretty sure I ran
> X-Windows (under Windows 3.1 !) while I still had a 386. OK, that was a
> long time ago, maybe it was a 486, but hey, used 486 motherboards are all
> over the place gathering dust, and another 8M of ram shouldn't be hard
> to obtain. Especially old-style ram SIMMs. Oh, I see a problem, many of
> those were only 1 or 2M, and you'd need 4x 4M. Another suggestion, new
> motherboards with 120MHz Pentiums, etc. are pretty cheap right now.
> That would make Linux really scream.
>
>--
>Winfield Hill hi...@rowland.org _/_/_/ _/_/_/_/
Hi Wiinfried, you are right Linux IS fast on a pentium.
I do howver not see how you could ever have run Linux in win.31 !
Linux is a 32 bit operatiing system, and, just for the record, it has a window 3.1 emulator (wabi).
Anyway, I run it on a 486 DX 2 66 with 20MByte.
Xwindows is quite usable then.
In fact, since I have Linux, I never get win 95, no use for it, only cost money.
And because Linux is multiuser not even NT can come close.
mmm I have no financial interests, sorry Bill Gates, thats the way it is.
But still looking for a program for Linux that can view and make .CGM files....
J.P.
Winfield Hill
Jan Panteltje at pa...@pi.net says...
>
>> Dan, does it take more than 8M to run X-Windows? I'm pretty sure I ran
>> X-Windows (under Windows 3.1 !) while I still had a 386. OK, that was a
>> long time ago, maybe it was a 486, but hey, used 486 motherboards are all
>> over the place gathering dust, and another 8M of ram shouldn't be hard
>> to obtain. Especially old-style ram SIMMs. Oh, I see a problem, many of
>> those were only 1 or 2M, and you'd need 4x 4M. Another suggestion, new
>> motherboards with 120MHz Pentiums, etc. are pretty cheap right now.
>> That would make Linux really scream.
>
> Linux is a 32 bit operatiing system, and, just for the record, it has a
No, I ran X-windows on Win3.1 - the version from NCD. Note: DESQview
also had a version but I didn't try it.
> In fact, since I have Linux, I never get win 95, no use for it, only
> cost money. And because Linux is multiuser not even NT can come close.
> mmm I have no financial interests, sorry Bill Gates, thats the way it is.
> But still looking for a program for Linux that can view and make .CGM
> files....
Well, Linux is great, but sometimes one needs Windows to run programs
not yet available on Linux. How well does wabi work? I'd hate to have
to go back to a Win 3.1 emulator. I didn't even like Win 3.1!!
Mark Zenier
In article <slrn6broo2....@insitu.physnet.uni-hamburg.de>,
Daniel Haude <Daniel...@public.uni-hamburg.de> wrote:
>On 14 Jan 1998 13:06:13 GMT,
> Winfield Hill <hi...@rowland.org> wrote:
>> So the .PNG is indeed pretty slick!! Now, if we can just
>> get all our readers out there up to speed!
>I'm using the SLRN newsreader on a Linux machine. It doesn't handle
>attachments, so I'd have to save the message to a file, and then run some
>decoder program to extract the PNG file. What program could I use for
>that? Is there a PNG viewer that runs without X?
>
>(My computer is a 386 w/ 8MB Ram. No X-Windows, no Netscape. Maybe that
>explains my fondness of ASCII art ;-)
Yes, there are a set of graphics viewer programs that will use the
'svga' graphics driver library. (If your kernal and the shared libraries
are up to date, you can run one of these viewer programs and the virtual
console stuff still works. Most of the time, this is one of the very few
things that can bollux up Linux). On the sunsite archive, look in the
apps/graphics/viewers/svga directory. I've used both 'zgv' and 'grav'.
They do .jpg, .gif, .png, .bmp, and probably others I've not run across
yet. Get the newest version, .png and .bmp were added in later ones.
Mark Zenier mze...@eskimo.com mze...@netcom.com
Kevin McMurtrie
In article <34BC17...@isinurl.net>, Mike <ad...@isinurl.net> wrote:
- - -
>Isn't PNG great? It's free, which means companies don't have to pay
>royalties for GIF. The decode is no more difficult than GIF, JPEG, or the
>rest of the compressed formats.
>
>A large number of DOS and Windows graphics viewers can read it.
>
>But none of the browsers I tried can :(
>
>Best Regards,
>
Sorry about being off-topic.
Installing QuickTime 3.0 (Windows & Mac) lets you open just about any format that can be turned into pixels or sound, including PNG.
Mike
QuickTime is 2,379 kbytes to download, possibly twice that after
installation. Then, you got to get all those QuickTime movies now that
you have the player - more time wasted on huge downloads.
My DOS MPEG player is 300k, and QPV takes 200k. These handle just about
any pixel-based graphics format I run in to, plus a lot of the
vector-based formats.
I simply cannot understand the reason for multi-megabyte programs!
P.S. Sorry for sending you the email also, Kevin - I hit the wrong button
Luiz Marcondes
On Sat, 17 Jan 1998 23:02:24 -0800, Mike wrote:
>Kevin McMurtrie wrote:
[snip]
>> Installing QuickTime 3.0 (Windows & Mac) lets you open just about any
>> format that can be turned into pixels or sound, including PNG.
>>
>> http://www.apple.com/quicktime/
>QuickTime is 2,379 kbytes to download, possibly twice that after
>installation. Then, you got to get all those QuickTime movies now that
>you have the player - more time wasted on huge downloads.
>My DOS MPEG player is 300k, and QPV takes 200k. These handle just about
>any pixel-based graphics format I run in to, plus a lot of the
>vector-based formats.
>I simply cannot understand the reason for multi-megabyte programs!
IrfanView32 (freeware!) is a small (260k) and fast loading package that
handle JPG, GIF, BMP, TIF, PNG, PCX, TGA, PCD, RAS, RLE, DIB, ICO, WMF and
EMF graphic formats, as well as AVI (video for windows) files.
It is a windows 95/NT program, and one of its niceties is that you can
easily set it as the default viewer for one/several/all the formats above.
On the other side, IrfanView will not allow more than one instance of itself
to run, meaning that if you click in several graphic files in Explorer, it
will show only the last. This can be a good thing if you are using IrfanView
as a "helper" program for a www browser, so you will not end with a lot of
graphic windows open after surfing the web, but sometimes I want to compare
two images and IrfanView doesn't allow me to do this... :(
Regards,
Luiz Marcondes
--
<yzn...@fhcre.mvccb.pbz> Apply ROT13 to decode
(Sorry by the inconvenience, avoiding spam...)
PGP key available