S
server
Guest
message unavailable
URGENT !! A nuclear plant near Minneapolis in th e U.S. state of Minnesota has leaked radioactive material fo r a second...
M
Martin Brown
Guest
On 01/04/2023 17:37, John Larkin wrote:
If you want to put some explanatory text info blurb on the front you can
just do it as is provided that there are no commas in the text. If there
are then enclose the thing in double quotes.
A leading single quote \' will act as a comment delimiter for the rest of
the line (but beware of commas).
Better control can be had by including all strings in double quotes \"
and escaping in any double quotes as \"\".
CSV might not be the most user friendly choice though these days. Tab
delimited is one of the Excel defaults and that is much more readable on
a dumb terminal. That said I more often see space delimited files.
Having text and column headings in quotes also makes the CSV/tabbed
files compatible with the likes of Gnu plot.
---sample of most common forms of text inclusion and mistakes ---
\' Full line comment, and contains escaped double quotes \"and works OK\"
\"Heading1\",\"heading2\",\"heading3\" \'
1.2 2.4 5678 or
1.2,2.4,5678,\"which would you prefer?\"
This is plain text on its own
\"This contains \"\"quoted\"\" text\"
This might not do what you expect, or then it might
--end--
Columns in tab delimited form rather than csv. I find it much easier for
browsing data files. Size is the same but readability on a dumb terminal
or in notepad or nano greatly enhanced.
One other useful trick is that in most cases ,, can be used as a
shorthand for 0, although the cell will be blank rather than 0.
(empty cells evaluate to numerical zero). It can be a huge saving in
image photon counting files where the majority of cells are zero.
--
Martin Brown
If you want to put some explanatory text info blurb on the front you can
just do it as is provided that there are no commas in the text. If there
are then enclose the thing in double quotes.
A leading single quote \' will act as a comment delimiter for the rest of
the line (but beware of commas).
Better control can be had by including all strings in double quotes \"
and escaping in any double quotes as \"\".
CSV might not be the most user friendly choice though these days. Tab
delimited is one of the Excel defaults and that is much more readable on
a dumb terminal. That said I more often see space delimited files.
Having text and column headings in quotes also makes the CSV/tabbed
files compatible with the likes of Gnu plot.
---sample of most common forms of text inclusion and mistakes ---
\' Full line comment, and contains escaped double quotes \"and works OK\"
\"Heading1\",\"heading2\",\"heading3\" \'
1.2 2.4 5678 or
1.2,2.4,5678,\"which would you prefer?\"
This is plain text on its own
\"This contains \"\"quoted\"\" text\"
This might not do what you expect, or then it might
--end--
Columns in tab delimited form rather than csv. I find it much easier for
browsing data files. Size is the same but readability on a dumb terminal
or in notepad or nano greatly enhanced.
One other useful trick is that in most cases ,, can be used as a
shorthand for 0, although the cell will be blank rather than 0.
(empty cells evaluate to numerical zero). It can be a huge saving in
image photon counting files where the majority of cells are zero.
--
Martin Brown
C
Carlos E.R.
Guest
On 2023-04-02 19:53, Don Y wrote:
You can figure out a way to do that without columns ;-)
....
Infinite ways. As text, for example. As a huge number.
--
Cheers, Carlos.
You can figure out a way to do that without columns ;-)
....
Infinite ways. As text, for example. As a huge number.
--
Cheers, Carlos.
J
Jan Panteltje
Guest
On Sun, 02 Apr 2023 16:26:53 GMT, Jan Panteltje <alien@comet.invalid
wrote:
Somehow every time I think about this
that thing I did see back in the fifties had a big tank with salt? water
and some metal plates put into it to control speed of things at some fair
comes to mind...
PS
I remember somebody doing a light dimmer for the school band
by de-capping a fluorescent light tube, filling it with salty water IIRC,
and lowering a wire with a piece of iron in it
using it as high power variable resistor.
Worked very nicely.
So, no electronics, but then you could use AI controlling a motor...
no end to computah power needed....
AI powered robot, can do the design too...
where will it go? .. After WW3 that is of course...
!
J
Jasen Betts
Guest
On 2023-04-01, Lamont Cranston <amdx62@gmail.com> wrote:
if your holes are large enough you could try desoldering tubes.
https://www.aliexpress.com/item/1005004200362252.html
they have a wall thickness of about 0.1mm so the leads need to be a
bit smaller than the hole.
The price suggests the build quality fairly accurately, you could
probably do better with a dispensing needle on a 10ml syringe body,
if you can find one the right size - needle needs to fit over the pin
and go through the hole.
--
Jasen.
ðºð¦ Слава УкÑаÑнÑ
if your holes are large enough you could try desoldering tubes.
https://www.aliexpress.com/item/1005004200362252.html
they have a wall thickness of about 0.1mm so the leads need to be a
bit smaller than the hole.
The price suggests the build quality fairly accurately, you could
probably do better with a dispensing needle on a 10ml syringe body,
if you can find one the right size - needle needs to fit over the pin
and go through the hole.
--
Jasen.
ðºð¦ Слава УкÑаÑнÑ
M
Martin Brown
Guest
On 01/04/2023 17:53, John Larkin wrote:
The cells of a spreadsheet are not \"normally numeric\" they are capable
of taking on any value that the computer can represent.
A few lines of text at the top with no commas in will be fine.
Your problem arises when something else expects the first line of the
file to contains headings in some particular format (eg Gnu plot).
They are extremely powerful tools for mocking up test data and having a
quick look see at modest amounts of experimental data.
The best thing about Excel or any other spreadsheet or scratchpad for
testing algorithms is that mistakes made in a spreadsheet are completely
orthogonal to those made in a procedural programming language.
You also have the oddity of Excel computing in FP and rounding to
strictly 15 decimal digits for display. This can sometimes cause
annoyance if you are working to machine precision. I have a work around
found by trial and error when I noticed an unusual value one day...
An example to demonstrate this behaviour (staring in A1):
column A column B column C
=1-2^ROW() =1-A1 =(1-A1)
And drag them down to row 55
You will see that column B is rounded to 15 decimal digits but column C
is full machine precision 53 mantissa. This can be very useful. But it
is quite arcane and not so far as I know documented Excel behaviour.
But then MS the help system would define \"Help\" as
\"He\" the masculine pronoun
\"LP\" obsolete black vinyl music disk now making a comeback.
--
Martin Brown
The cells of a spreadsheet are not \"normally numeric\" they are capable
of taking on any value that the computer can represent.
A few lines of text at the top with no commas in will be fine.
Your problem arises when something else expects the first line of the
file to contains headings in some particular format (eg Gnu plot).
They are extremely powerful tools for mocking up test data and having a
quick look see at modest amounts of experimental data.
The best thing about Excel or any other spreadsheet or scratchpad for
testing algorithms is that mistakes made in a spreadsheet are completely
orthogonal to those made in a procedural programming language.
You also have the oddity of Excel computing in FP and rounding to
strictly 15 decimal digits for display. This can sometimes cause
annoyance if you are working to machine precision. I have a work around
found by trial and error when I noticed an unusual value one day...
An example to demonstrate this behaviour (staring in A1):
column A column B column C
=1-2^ROW() =1-A1 =(1-A1)
And drag them down to row 55
You will see that column B is rounded to 15 decimal digits but column C
is full machine precision 53 mantissa. This can be very useful. But it
is quite arcane and not so far as I know documented Excel behaviour.
But then MS the help system would define \"Help\" as
\"He\" the masculine pronoun
\"LP\" obsolete black vinyl music disk now making a comeback.
--
Martin Brown
D
Don Y
Guest
On 4/3/2023 1:08 AM, Carlos E.R. wrote:
Not in a spreadsheet. Unless you treat the spreadsheet as a text table
and all of the data as typeless.
In which case, why aren\'t you just using a word processor? And, living
with the inevitable mistakes that will creep into your data? E.g., l00
is not a number -- despite looking like one!
In a database (\"relation\") you can \"relate\" different tables to each other.
So, in addition to 1:1 relationships (implicit in a spreadsheet), you can
have 1:1, 1:n, n:m, n:1, etc. relationships, as dictated by the data.
The reason you want \"typed\" fields is to reduce errors.
Is this a number: THIS
Is this a MAC: HeLlOboysandgirls
Is this a float: 123.456.78,,90
DBMSs present so many richer ways of presenting and preserving
data to make the DBAs job easier.
What do you do when there are two entries in a spreadsheet that
obviously refer to the same thing? E.g., a part number having
two different descriptions? In a DBMS, you can ensure that
all part numbers are unique (trying to add another record with
an existing part number throws an error). You can also encode
other criteria to safeguard the data going into the relation.
E.g., \"biological child\'s age must be less than age of parents\"
(duh!) or \"birthdate must be in the past\", etc.
Try some simple problems in a spreadsheet... like keeping track
of the songs in your music collection. Then, try to convince
yourself that your spreadsheet is better than just a textual table!
Or, document your family tree for a couple of generations.
Not in a spreadsheet. Unless you treat the spreadsheet as a text table
and all of the data as typeless.
In which case, why aren\'t you just using a word processor? And, living
with the inevitable mistakes that will creep into your data? E.g., l00
is not a number -- despite looking like one!
In a database (\"relation\") you can \"relate\" different tables to each other.
So, in addition to 1:1 relationships (implicit in a spreadsheet), you can
have 1:1, 1:n, n:m, n:1, etc. relationships, as dictated by the data.
The reason you want \"typed\" fields is to reduce errors.
Is this a number: THIS
Is this a MAC: HeLlOboysandgirls
Is this a float: 123.456.78,,90
DBMSs present so many richer ways of presenting and preserving
data to make the DBAs job easier.
What do you do when there are two entries in a spreadsheet that
obviously refer to the same thing? E.g., a part number having
two different descriptions? In a DBMS, you can ensure that
all part numbers are unique (trying to add another record with
an existing part number throws an error). You can also encode
other criteria to safeguard the data going into the relation.
E.g., \"biological child\'s age must be less than age of parents\"
(duh!) or \"birthdate must be in the past\", etc.
Try some simple problems in a spreadsheet... like keeping track
of the songs in your music collection. Then, try to convince
yourself that your spreadsheet is better than just a textual table!
Or, document your family tree for a couple of generations.
M
Martin Brown
Guest
On 01/04/2023 15:00, Fred Bloggs wrote:
They might just possibly be communicating with insects - saying words to
the effect of \"Go away - I\'m toxic\". But I don\'t see an obvious
mechanism for them to generate such ultrasonic clicks though.
Capillary effects in xylem are strong enough to support very tall
columns of water as tall trees demonstrate so well.
--
Martin Brown
They might just possibly be communicating with insects - saying words to
the effect of \"Go away - I\'m toxic\". But I don\'t see an obvious
mechanism for them to generate such ultrasonic clicks though.
Capillary effects in xylem are strong enough to support very tall
columns of water as tall trees demonstrate so well.
--
Martin Brown
C
Chris Jones
Guest
On 2/04/2023 2:41 am, Lamont Cranston wrote:
I have some wide tips for my Metcal e.g. SMTC-161 (though sometimes a
less wide one like SMTC-160 fits better). Those can melt all of the
joints on one side of a DIP package or two pins on a switch, and then I
can tilt it and get that side further from the board. Then I let it cool
before releasing the force, and melt the other side and tilt it the
other way, let it cool, and then go back and do the other side. By
rocking it back and forth I can get it out without damaging anything.
A good solder sucker might also work.
If there are internal ground planes without thermal reliefs then all
bets are off, and the big hot air gun is usually needed, sometimes in
addition to the metcal.
I have some wide tips for my Metcal e.g. SMTC-161 (though sometimes a
less wide one like SMTC-160 fits better). Those can melt all of the
joints on one side of a DIP package or two pins on a switch, and then I
can tilt it and get that side further from the board. Then I let it cool
before releasing the force, and melt the other side and tilt it the
other way, let it cool, and then go back and do the other side. By
rocking it back and forth I can get it out without damaging anything.
A good solder sucker might also work.
If there are internal ground planes without thermal reliefs then all
bets are off, and the big hot air gun is usually needed, sometimes in
addition to the metcal.
L
Lamont Cranston
Guest
OP here, I\'m surprised this is getting so much attention. I felt a little silly even asking the question,
about desoldering a switch. The switches I removed came out very easy. Not what I expected! I did crank up
the temp on my soldering iron to 830* and just moved from pin to pin with a little prying force near each pin with
an Exacto blade, until I got enough room to insert my needle nose pliers and then used those to pry with.
After the switches were out I added solder and then heated the topside and sucked out the solder from the other side.
The holes were very clean after that, didn\'t have to do any further solder removal. This was my Arduino
and they are fairly inexpensive. Recently my son brought a pcb home from work that was some type of controller,
it had a bad membrane switch in the middle of the board, surrounded by parts, not any room to get a
good prying action, I didn\'t want to attempt removal on that board. It wasn\'t mine, it was expensive and
would stop work until they got a new one, if they could. We just added wires and outboarded a new switch.
There was plenty of room on the front panel for adding the outboarded switch.
Mikek
about desoldering a switch. The switches I removed came out very easy. Not what I expected! I did crank up
the temp on my soldering iron to 830* and just moved from pin to pin with a little prying force near each pin with
an Exacto blade, until I got enough room to insert my needle nose pliers and then used those to pry with.
After the switches were out I added solder and then heated the topside and sucked out the solder from the other side.
The holes were very clean after that, didn\'t have to do any further solder removal. This was my Arduino
and they are fairly inexpensive. Recently my son brought a pcb home from work that was some type of controller,
it had a bad membrane switch in the middle of the board, surrounded by parts, not any room to get a
good prying action, I didn\'t want to attempt removal on that board. It wasn\'t mine, it was expensive and
would stop work until they got a new one, if they could. We just added wires and outboarded a new switch.
There was plenty of room on the front panel for adding the outboarded switch.
Mikek
J
Jasen Betts
Guest
On 2023-03-31, Sid 03 <sidwelle@gmail.com> wrote:
NFPA NEC
> And why can\'t I find a chart that follows the AWG down into the 2? AWG gauges ?
NEC doesn\'t allow LV on those sizes.
Figure out how much energy you can afford to lose in the wires and choose a
size that does better
--
Jasen.
ðºð¦ Слава УкÑаÑнÑ
NFPA NEC
> And why can\'t I find a chart that follows the AWG down into the 2? AWG gauges ?
NEC doesn\'t allow LV on those sizes.
Figure out how much energy you can afford to lose in the wires and choose a
size that does better
--
Jasen.
ðºð¦ Слава УкÑаÑнÑ
F
Fred Bloggs
Guest
On Sunday, April 2, 2023 at 7:14:17â¯PM UTC-4, a a wrote:
https://www.fema.gov/emergency-managers/risk-management/safe-rooms/protection
People need a room they can walk into like any other room in their house. Those underground bomb shelter type rooms requiring going outside and climbing, descending actually, a ladder don\'t cut it for ultra-short notice events or people with mobility issues. These are very expensive, on order of $1000/sf. Mainly walls have to hold up against being struck by objects travelling 300 MPH, and the substructure has to be anchored with 10s of thousands lbs of weight ( thick concrete slab w/strong anchoring )- tornadoes have no problem uprooting big trees means they have no problem making your safe room go airborne. It can be done, DIY actually.
https://www.forbes.com/sites/quora/2016/03/22/what-would-it-take-to-build-a-completely-tornado-proof-house/?sh=627b29ed27ca
You don\'t need to make the WHOLE house tornado proof to survive- cost and aesthetically prohibitive.
https://www.fema.gov/emergency-managers/risk-management/safe-rooms/protection
People need a room they can walk into like any other room in their house. Those underground bomb shelter type rooms requiring going outside and climbing, descending actually, a ladder don\'t cut it for ultra-short notice events or people with mobility issues. These are very expensive, on order of $1000/sf. Mainly walls have to hold up against being struck by objects travelling 300 MPH, and the substructure has to be anchored with 10s of thousands lbs of weight ( thick concrete slab w/strong anchoring )- tornadoes have no problem uprooting big trees means they have no problem making your safe room go airborne. It can be done, DIY actually.
https://www.forbes.com/sites/quora/2016/03/22/what-would-it-take-to-build-a-completely-tornado-proof-house/?sh=627b29ed27ca
You don\'t need to make the WHOLE house tornado proof to survive- cost and aesthetically prohibitive.
F
Fred Bloggs
Guest
On Monday, April 3, 2023 at 6:32:52â¯AM UTC-4, Martin Brown wrote:
Pretty sure they were talking about the herbaceous forb variety, not ligneous.
Pretty sure they were talking about the herbaceous forb variety, not ligneous.
A
a a
Guest
On Monday, 3 April 2023 at 15:40:53 UTC+2, Fred Bloggs wrote:
Another fake research paper
R
Ralph Mowery
Guest
In article <u0ega9$6dl$1@gonzo.revmaps.no-ip.org>, usenet@revmaps.no-
ip.org says...
There are 2 main things to look at for wire sizes. The amount of
voltage drop and the temperature rise.
Not lookimg at any codes , but an example. You could use # 16 wire for
say a 10 foot dropcord and only loose a couple of volts. Extend it out
to 100 or 200 feet and the same load would drop say 15 volts. So while
it would carry the curent the voltage drop would be excessive and you
would need to go up to # 12 or # 10.
For temperature rise it would depend on the insulation type and if it
was in open air or in a conduit with several similar wires or wound in a
transformer .
ip.org says...
There are 2 main things to look at for wire sizes. The amount of
voltage drop and the temperature rise.
Not lookimg at any codes , but an example. You could use # 16 wire for
say a 10 foot dropcord and only loose a couple of volts. Extend it out
to 100 or 200 feet and the same load would drop say 15 volts. So while
it would carry the curent the voltage drop would be excessive and you
would need to go up to # 12 or # 10.
For temperature rise it would depend on the insulation type and if it
was in open air or in a conduit with several similar wires or wound in a
transformer .
B
Bob Engelhardt
Guest
On 4/3/2023 4:07 AM, Jasen Betts wrote:
Well, I\'ll be ... as they used to say. I\'ve never heard of them, but
they look like they could be very useful.
for a coupla\' bux it\'s worth the chance that the more expensive ones are
actually better.
Thanks for the tip.
Well, I\'ll be ... as they used to say. I\'ve never heard of them, but
they look like they could be very useful.
I dunno\' ... the pictures and descriptions are exactly the same. But
for a coupla\' bux it\'s worth the chance that the more expensive ones are
actually better.
Thanks for the tip.
B
Bob Engelhardt
Guest
On 4/2/2023 2:54 PM, Fred Bloggs wrote:
is slimy - very hard to grab & keep a hold of.
Yet, I expect very little to come of it. He\'s not so much Teflon as he
is slimy - very hard to grab & keep a hold of.
A
a a
Guest
>
Yet one more #veryStupidByLowIQaa post.
Yet one more #veryStupidByLowIQaa post.
J
John Larkin
Guest
On Mon, 03 Apr 2023 05:43:58 GMT, Jan Panteltje <alien@comet.invalid>
wrote:
I can compute the power dissipation of each channel, based on the
programmed resistance and digitized voltage and current. Air flow is
left to right on the board, but it could fail too and the board would
cook. So we need a surface temperature sensor or two.
This goes in a chassis with up to 8 plugin cards and two big
controlled fans. My idea is for each board to present one bit to the
main controller, I\'M TOO HOT. If the controller sees any such bit from
any board, it nudges up fan speeds. If no bits, it nudges down towards
minimum. Maybe 2% fan speed per second?
A user will send a SCPI command to select a resistance, which will set
up a solid-state-relay pattern. That won\'t happen often and needn\'t be
fast. The SSRs themselves operate in roughly a millisecond.
wrote:
On a sunny day (Sun, 02 Apr 2023 10:08:11 -0700) it happened John Larkin
jlarkin@highlandSNIPMEtechnology.com> wrote in
fmcj2ipm5so0l2obinqami6lc38amc92ap@4ax.com>:
On Sun, 02 Apr 2023 16:26:53 GMT, Jan Panteltje <alien@comet.invalid
wrote:
On a sunny day (Sun, 02 Apr 2023 08:00:00 -0700) it happened John Larkin
jlarkin@highlandSNIPMEtechnology.com> wrote in
ak5j2i13tp7qtgf113cfor3oslpppbndv0@4ax.com>:
On Sun, 02 Apr 2023 14:32:40 GMT, Jan Panteltje <alien@comet.invalid
wrote:
On a sunny day (Sun, 2 Apr 2023 16:14:53 +0200) it happened \"Carlos E.R.\"
robin_listas@es.invalid> wrote in <t7fofjxe1d.ln2@Telcontar.valinor>:
On 2023-04-02 15:23, Jan Panteltje wrote:
On a sunny day (Sun, 2 Apr 2023 14:13:05 +0200) it happened \"Carlos E.R.\"
robin_listas@es.invalid> wrote in <h38ofjx1t.ln2@Telcontar.valinor>:
On 2023-04-02 07:50, Jan Panteltje wrote:
On a sunny day (Sat, 1 Apr 2023 19:55:36 -0700 (PDT)) it happened Simon S
Aysdie <gwhite@ti.com> wrote in
9a5cb22f-c985-45c7-914b-38adb0287aa4n@googlegroups.com>:
On Saturday, April 1, 2023 at 9:11:05â¯AM UTC-7, John Larkin wrote:
I\'m writing a program to create a CSV file that lists possible
resistor matrix values. I might hand it off to other people to use. I
don\'t use spreadsheets so don\'t know much about them.
I can do lines with numbers and strings and commas, but I\'d like the
spreadsheets to start with introductory comments. So, how does one
format comments into a csv file?
A CSV file isn\'t a spread spreadsheet. It is a text file. Put your introduc=
tory strings in separated by commas.
You can put something like a # to start a line/comment, if you want. Your i=
nterpreter needs to be written to understand it.
--------
Make a text file called text.csv with a text editor and then open it with e=
xcel.,,,\" \"
2A,,,2D
,,3C,\" \"
,4B,,\" \"
5A,,,5D
closing,,comment,\" \"
--------
Above, I am not sure if the\\t in between the \'\" will make it past gaggle gr=
oups. (I mean \"\\t\")
Hey, that TAB appears here as 8 spaces
Could be 4 spaces too, depends?
Just using spaces should be OK too I think..
I was wondering what happens if you use European notition for numbers..
300,2 for 300.2 for example ;-)
You can use semicolons as separators instead of commas ;-)
I did write some code to allow input like 1,456,789,911
in the frequency input fields of my spectrum analyzer...
https://panteltje.nl/pub/xpsa_fm_spectrum.gif
If he has numbers like that for that spreadsheet then he may encounter that,
csv format does not make a lot of sense for numbers.
It should work, setting the locale appropriately (USA). I just tried with:
cer@Telcontar:~> cat p.csv
1;2;3;1,456,789,911;\"coment\"
2;2;3;1,456,789,911;\"coment\"
3;2;3;1,456,789,911;\"coment\"
4;2;3;1,456,789,911;\"coment\"
5;2;3;1,456,789,911;\"coment\"
6;2;3;1,456,789,911;\"coment\"
7;2;3;1,456,789,911;\"coment\"
8;2;3;1,456,789,911;\"coment\"
cer@Telcontar:~> localc p.csv
And it worked. If I leave locale as Spain, it doesn\'t.
OK, nice.
Yes Spain / Europe will see 2 or more commas in a field as wrong.
And ,911 I deliberately used will alarm the NSA ;-)
Testing mike:
1,2,3,4,5,6,7,8,9,11
Hers\'s my file:
https://www.dropbox.com/s/y8pkfpmjhcs3aml/P955_m2.csv?dl=0
I have a matrix of resistors with 9 programmable SSRs shorting various
nodes. I want a range of resistances that users can select, 32
different values maybe, plus open, short, and ground fault.
Some resistances are redundant (or very close) but have different
voltage
should add that to the math and then maybe sort by resistance to help
people select.
Interesting little problem. The design is pretty dull otherwise.
https://www.dropbox.com/s/qh6rxnhaujq8n29/M2.jpg?raw=1
Well, cannot get much info from that PCB ? Circuit?
resistors 1, 2, 4, 8, 16 etc?
9 bits, 512 values matches your list
The little bits (high values) would be lower power ones?
The software a lookup list of the customer selector knob to binary?
Here\'s my current thinking. I started with just a conductance DAC,
namely resisors R, 2R, 4R etc in parallel, each with an SSR to switch
them in and out.
https://www.dropbox.com/s/3jpo8n6s7wrhnsl/Matrix_m2.jpg?raw=1
It gets complicated for the high bits
The lower 3 bits OK.
This lets me put 5-watt resistors in series and parallel, which helps
with power dissipation. And adds \"short\" for free.
Interesting solution, really!
We\'d prowl the spreadsheet and pick an assortment of resistance values
that the customer can request. A lookup table would map them to relay
patterns.
Yes
Of course a customer can program some resistance and then apply some
outrageous voltage and blow things up. The \"short\" case needs
protection too. I\'m optimistic that if I digitize voltage and current,
an FPGA can shut things off before we fry resistors or SSRs. We\'ll
need an isolated 2-channel ADC per load.
Maybe a simple hardware comparator across that .1 Ohm series shunt?
Should be faster?
Same for input voltage...
Thermal protection too..
I can compute the power dissipation of each channel, based on the
programmed resistance and digitized voltage and current. Air flow is
left to right on the board, but it could fail too and the board would
cook. So we need a surface temperature sensor or two.
This goes in a chassis with up to 8 plugin cards and two big
controlled fans. My idea is for each board to present one bit to the
main controller, I\'M TOO HOT. If the controller sees any such bit from
any board, it nudges up fan speeds. If no bits, it nudges down towards
minimum. Maybe 2% fan speed per second?
A user will send a SCPI command to select a resistance, which will set
up a solid-state-relay pattern. That won\'t happen often and needn\'t be
fast. The SSRs themselves operate in roughly a millisecond.
Somehow every time I think about this
that thing I did see back in the fifties had a big tank with salt? water
and some metal plates put into it to control speed of things at some fair
comes to mind...
J
John Larkin
Guest
On Mon, 03 Apr 2023 08:14:32 GMT, Jan Panteltje <alien@comet.invalid>
wrote:
We once used a full-sized plastic garbage can, full of water, as a
dummy load, and we boiled it.
wrote:
On Sun, 02 Apr 2023 16:26:53 GMT, Jan Panteltje <alien@comet.invalid
wrote:
Somehow every time I think about this
that thing I did see back in the fifties had a big tank with salt? water
and some metal plates put into it to control speed of things at some fair
comes to mind...
PS
I remember somebody doing a light dimmer for the school band
by de-capping a fluorescent light tube, filling it with salty water IIRC,
and lowering a wire with a piece of iron in it
using it as high power variable resistor.
Worked very nicely.
So, no electronics, but then you could use AI controlling a motor...
no end to computah power needed....
AI powered robot, can do the design too...
where will it go? .. After WW3 that is of course...
!
We once used a full-sized plastic garbage can, full of water, as a
dummy load, and we boiled it.