Driver to drive?

[Ban]

Spehro Pefhany wrote:

That's why I wrote 'discomfort'. my own laptop does get uncomfortably
warm, but it doesn't burn me.
snip

http://www.smh.com.au/articles/2002/11/22/1037697857595.html


I suspect in this case the CPU was from a desktop and not a low power Mobile

Pentium meant to be for laptops. And probably that guy felt a cosy heat,
after all it was in Sweden in autumn. And maybe those cloths covered the
cooling fan openings.
What a "scientist", carried away by his mind he ruins his private parts.
)
What a nerd.

--
ciao Ban
Bordighera, Italy

 

[Bob Stephens]

On 6 Oct 2004 08:30:49 -0700, sarah wrote:

Hi,

I run protel dxp compile PCB project and got the warning: off-grid
netlist label ... I am wondering what does it mean and how to solve
it.

Thanks.

Sarah

I get those sorts of errors if I place something with the grid set to 1 mil
say, and then change the grid to 5 or 10 mils. The compiler doesn't like
coordinates that aren't integer multiples of the grid. Don't know if this
is what you're seeing or not. As regards connectivity, I've found that
DXP/2004 is much more tolerant of slightly misaligned labels than previous
versions of Protel.

Bob

 

[Mark Zenier]

In article <1fb4m0la9abpqbcn8cmkv1pmjjfahjjthi@4ax.com>,
Jonathan Kirwan <jkirwan@easystreet.com> wrote:

On 4 Oct 2004 21:42:04 -0700, kenkeeley@hotmail.com (Kenneth Keeley) wrote:

I have looked around the web but haven't found any well shown designs
all of the ones I have found have talked about using a PIC or micro
controller to interface with the keyboard.

That should help, I'd imagine. But it doesn't give you the theory, I suppose.

Do you know where I could
find the IBM docs that you referred to.

Yeah. On my shelf, in the sometimes coveted, 4-volume set called the "IBM
Technical Reference." I'd just figured that someone out there must have
documented all the important details, already.

It was in a couple of the columns in Circuit Cellar Ink magazine, back
about 10 years ago. Both Ed Nisley and John Dybowski did keyboard
interfaces, and Ed Nisley's stuff talked about all three modes that
an AT keyboard would run. (Stuff I'd not seen anywhere else).

Nisley did that as part of a series on using the PC in embedded
applications, I think it got made into a book.

There's an index on www.circuitcellar.com, as I've heard.

Mark Zenier mzenier@eskimo.com Washington State resident

 

[Phil Hobbs]

colin wrote:

Amplifiers like to oscilate whenever they get the chance, its what makes
analogue electronics fun, its usualy some sneak positve feedback path via
the gnd or suply traces, or negative feedback that has overal phase shift of
180 at greater than unity gain. its hard to say without seeing the circuit,
im not exactly familiar with class H tho, i only know the power suply is
modulated with the signal (maybe with a smps ?), wich leads to a wide range
of posibilities for oscilation cuases.

Of course, the way to make an amplifier stable is to try making an
oscillator.

Cheers,

Phil Hobbs

 

[Rene Tschaggelar]

On the 9 pin connector of the PC, the signals are :
Tx on 3, Rx on 2, GND on 5

On the 25 pin connector of the PC, the signals are :
Tx on 2, Rx on 3, GND on 7

Rene
--
Ing.Buero R.Tschaggelar - http://www.ibrtses.com
& commercial newsgroups - http://www.talkto.net


Boki wrote:

Deaer Roger,
Yes, I connected pin.3(TXD) to oscilloscope signal and
pin.1 connected to oscilloscope ground.(didn't swap lines)
but there is no signal in oscilloscope. If I short pin.2 pin3, it works.

I am confusing why it didn't send out signals when pin.2 pin.3 are not
short. ( I floting all other pins.,.except signal and ground.)

Best regards,
Boki.

 

[Rich Grise]

On Monday 04 October 2004 01:31 am, John Woodgate did deign to grace us with
the following:

I read in sci.electronics.design that Rich Grise <null@example.net
wrote (in <pc28d.3805$na.1576@trnddc04&gt about 'Marketing blurb -
bullsh*t baffles etc', on Mon, 4 Oct 2004:
On Sunday 03 October 2004 11:57 am, John Woodgate did deign to grace us
with the following:

I read in sci.electronics.design that Rich Grise <null@example.net
wrote (in <3CX7d.4845$x65.3003@trnddc06&gt about 'Marketing blurb -
bullsh*t baffles etc', on Sun, 3 Oct 2004:

Psychrotronic!

Imitation is said to be the sincerest form of flattery. But you don't
know what it means.
--
It means something?

Yes, it was coined a very long time ago by columnist 'Free Grid' in
'Wireless World', a classical scholar. It means 'cold electronic
device', from Greek 'psychro' = cold, and '-tron' from 'elektron' -
amber. In fact he was speculating about tubes with field-emission
cathodes, long before the concept appeared in other than speculative
circles.
--

And here, I thought I'd made it up from random fluctuations in my quantum
soup.

But according to this, I've received telepathic communication! I had
never heard the word before, in fact, I thought I made it up, when clearly,
I received it telepathically from the metaphysical quantum foam.

Thanks!
Rich

 

[Roger Hamlett]

"Boki" <bokiteam@ms21.hinet.net> wrote in message
news:ck14g8$bfl$1@netnews.hinet.net...

Deaer Roger,
Yes, I connected pin.3(TXD) to oscilloscope signal
and
pin.1 connected to oscilloscope ground.(didn't swap lines)
but there is no signal in oscilloscope. If I short pin.2 pin3, it works.

I am confusing why it didn't send out signals when pin.2 pin.3 are not
short. ( I floting all other pins.,.except signal and ground.)

Best regards,
Boki.
OK.

One thought, is that if the receive input is disconnected, the code will
be seing a continuous 'break'. It may be that this is disabling the
transmission. Try just pulling the RX line low, and then seeing if TX
starts. The handshake lines should have a suitable signal available (RTS
or CTS).

Best Wishes

 

[John Larkin]

On Wed, 06 Oct 2004 09:31:01 -0400, Spehro Pefhany
<speffSNIP@interlogDOTyou.knowwhat> wrote:

On 06 Oct 2004 12:51:22 GMT, the renowned chrisgibbogibson@aol.com
(ChrisGibboGibson) wrote:

Also, originally, after a reading, the ramp was left to run high and I noticed
that this affected the following reading quite dramatically. Ensuring it always
runs to the same height made a big improvement.

BTW, I've used this effect to advantage in a commercial product to
deliberately de-linearize an ADC to match a certain sensor's (slight)
nonlinearity. It works okay for small amounts of nonlinearity relative
to the desired accuracy of the ADC, but the tolerances of inexpensive
capacitors are nothing to write home about, so it's probably not as
applicable if you need something really nonlinear. Essentially I made
an artificial capacitor with huge DA.

You can do the reverse: feed back a bit of the ramp voltage into the
guts of the current source such as to increase the current as the ramp
runs. This adds a t-squared term to the curve, nearly compensating
shunt resistance or any other effects that bow the ramp down. I've
done this with 25 ns ramps.

Reminds me of the old RTD linearization trick. Well, sort of.

John

 

[Charles Edmondson]

ranmyaku wrote:

I am looking for a way to control small voltages (0-5v) remotly using
my PC and WiFi if possible. Does anyone know of such a product, or of
how to build one mysef (writing software isn't a problem)? Thanks.

http://www.dpactech.com/


--
Charlie
--
Edmondson Engineering
Unique Solutions to Unusual Problems

 

[colin]

"Yannick" <yannick_de_wit@pandora.be> wrote in message
news:cc50d220.0410060820.3cd14173@posting.google.com...

you would conect it as per the bootstrap or comon base transistor, ie
gate
goes where the base is etc.
i havnt read the book but i asume its the same as in here
http://users.bestweb.net/~hobbs/frontends/frontends.pdf


Ok, soo with one mosfet like the cascoding circuit (common base) with
the gate at a constant voltage higher then VT and the drain and source
bewteen photodiode and transimpedance, right?

That sounds about right. the source would go where the emiter is etc.

there is usualy a diode between drain-source in a mosfet so u cant
normaly
flip them.


Ah , that's strange soo there must be some physical difference between
them (maybe parasitic capacitance wich will differ) because in theory
they are equal.

in theory lots of things are easier, however there is an unavoidable diode
from each end of the chanel to the substrate. some mosfets have this
substrate brought out on a 4th lead and then the D an S truly are
interchangaeble, however to leave it floatling wil cuase problems with drift
in gate voltage etc so its usualy tied to the source wich means under normal
operation the diode is reverse biased as an n chanel device is usualy
operated with drain positive.

basicaly becuase you can generaly chose a discrete transistor with lower
noise than the op amp, also a diferential amplifier has 2 transistors
wich
the noise adds together. the bootstrap and comon base transistor
provides
some gain therefore the signal is stronger before it gets to the op amp
so
the opamps noise wil not degrade the SNR so much.

Ofcourse, i am going to try this.

Im not sure how well this wil lend itself to high frequency circuits. The
results are ploted for up to about 1mhz wich is comparitvly quite low, with
a BJT its a qustion of selecting the bias curent so that the contribution of
noise curent and noise voltage is about equal, as frequency goes up the
impedance gets lower so the optimum bias curent wil go up, giving lower
noise voltage, im not sure how parasitic elements would afect it, it would
be interesting to see. with a mosfet the noise curent is usualy independant
of the drain curent, and noise voltage fals with increasing curent up to a
point where the device heats up so much it has more thermal noise i gues.
however fets tend to have higher capacitances wich would easily kill of any
benefit, unless you used a very high performance device. dont forget that to
get the high curents you would need a low value bias resistor wich might
kill the noise performance, or a curent source wich would need to be a very
low noise indeed.

However this is going in the oposite direction of my preference wich would
be for simpler stages. the more complications you add the more potential for
pitfalls. Im not sure if you would actualy be able to acehive the
improvement that it sugests. Im realy mentioning this as food for thought
rather than advice. i would see how noise is afecting the acuracy of the
phase measurments before adding even more complexity.

with 2 seperate stages the second stage is seeing a much stronger signal
so
its noise contribution isnt going to afect the SNR by much, also the
impedance would be fairly low so you could chose a very low voltage
noise
bipolar amp.


I was thinking of using the OPA847 for first stage after mosfet
because it has ultra low voltage noise : 0.8nv/rootHz , the current
noise is higher, i have to calculate wich is best... maybe the opa657
is better for a first stage, after this one i want to use a current
feedback opamp to provide very large gain over a wide frequency with
more current noise but this wont affect the SN ratio much due
increased signal. and so i can lower the gain for the lower
frequenties to maintain a flat response to the bandwidth i want to
have...

You can transfer the noise contribution of the 2nd amplifier back to the
first by dividing it by the gain of the first amplifier. ie if the first
amplifier has a voltage gain of 10 then the amplifier you mention wil
efectivly add 0.08 nv to the input wich is realy neglegible compared to the
4nv of the mosfet op amp.

Positive feedback has the oposite efect of negative feedback (obviously) ie
it increases the gain, but this isnt always free, too much and it will of
course become an oscilator.

If you need to amplify the signal to a significant level, another problem
you wil have to consider is the varition of phase delay with signal
strength, especialy if the amplifier becomes overloaded, wich can easily
hapen if you have enough gain to cope with the weakest of signals and then
receive a very strongly reflected signal, wich may drive the final amp in to
limiting. im using a FM ic in wich the last stage is a purpose designed
limiting amplifier.

Colin =^.^=

 

[Spehro Pefhany]

On Mon, 4 Oct 2004 15:37:05 GMT, the renowned "Ken Finney"
<kenneth.c.finney@boeing.com> wrote:

"Spehro Pefhany" <speffSNIP@interlogDOTyou.knowwhat> wrote in message
news:23m0m0h1bt4vbi5pmuicsn76cn73nbq6f9@4ax.com...
On Sun, 03 Oct 2004 17:50:23 GMT, the renowned Fred Bloggs
nospam@nospam.com> wrote:

Now USAID is
scrambling to salvage credibility- their position is hopeless- and
another US agency bites the dust.

http://www.guardian.co.uk/sudan/story/0,14658,1318643,00.html


Interesting. I don't watch US news, almost exclusively depending on
the BBC, DW, and Canadian news. BBC has been focusing on
Darfur for many months, and they have been yelling "genocide" since
the beginning.

Ah, well, you know the BBC has been thoroughly discredited. ;-)

I recently listened to a top diplomat (under the Chatham House Rule,
or I'd name names) tell us that things are going just swimingly in
Afghanistan, and when called on it by some VERY people who also are
there on the ground (and even leave Kabul at times), admitted and
defended his spin by saying that the other side has been covered well
by the media. I'm sure that the fact that this ambitious young fellow
handles hundreds of millions of taxpayer dollars in aid has little
effect on his opinions.


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[John Woodgate]

I read in sci.electronics.design that Rich Grise <null@example.net>
wrote (in <8wa8d.4133$eq1.724@trnddc08&gt about 'Marketing blurb -
bullsh*t baffles etc', on Mon, 4 Oct 2004:

And here, I thought I'd made it up from random fluctuations in my
quantum soup.

But according to this, I've received telepathic communication! I had
never heard the word before, in fact, I thought I made it up, when
clearly, I received it telepathically from the metaphysical quantum
foam.

No, you read it in one of my posts yesterday or the day before, that
translation from French, but perhaps it didn't register in your
conscious mind.
--
Regards, John Woodgate, OOO - Own Opinions Only.
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

 

[Mike Monett]

John Larkin wrote:

On Wed, 06 Oct 2004 09:31:01 -0400, Spehro Pefhany
speffSNIP@interlogDOTyou.knowwhat> wrote:

[...]

BTW, I've used this effect to advantage in a commercial product to
deliberately de-linearize an ADC to match a certain sensor's (slight)
nonlinearity. It works okay for small amounts of nonlinearity relative
to the desired accuracy of the ADC, but the tolerances of inexpensive
capacitors are nothing to write home about, so it's probably not as
applicable if you need something really nonlinear. Essentially I made
an artificial capacitor with huge DA.


You can do the reverse: feed back a bit of the ramp voltage into the
guts of the current source such as to increase the current as the ramp
runs. This adds a t-squared term to the curve, nearly compensating
shunt resistance or any other effects that bow the ramp down. I've
done this with 25 ns ramps.

Reminds me of the old RTD linearization trick. Well, sort of.

John

I have nothing but the highest regard for the two of you. Adding a
calibrated distortion to compensate for some nonlinearity demonstrates
the highest skill and knowledge.

But I have to wonder sometimes if it's worth it?? If you can get the
system response into digital form, it should be easy to generate a
polynomial or some other correction factor and eliminate the errors.

The advantage of this approach is the system can be calibrated
automatically in manufacturing, and can include other variations such as
temperature, humidity, supply voltage, or any other parameter of
interest. With the low cost of automated test systems and a bit of
software, it should pay for itself for manufacturing runs of two or more.

After the data is collected over the manufacturing run, it can be
analyzed to spot unusal variations that may indicate a potential failure,
or show where the performance can be improved, or where the cost can be
trimmed.

The first approach adds components to each system, and requires time and
effort to check the calibration for each system that is shipped. The
second approach does everything in software, eliminates as many parts as
possible, and gives a manufacturing database that can be used for further
improvements. The benefits of the second method are clear, but when would
it make sense to choose the first option?

Mike Monett

 

[John Fields]

On Wed, 06 Oct 2004 11:56:19 GMT, wouter@voti.nl (Wouter van Ooijen
(www.voti.nl)) wrote:

My question was about the symbol, not about chips. The chip designer
can of course choose what he wants (7473 and 7473A choose
differently!), I am interested in how it should be drawn as a symbol.

---
http://focus.ti.com/lit/ml/sdyz001a/sdyz001a.pdf

--
John Fields

 

[Jamie]

Wouter van Ooijen (www.voti.nl) wrote:

I am confused about the edge on which a FF triggers. My intuition
would say that any logic symbol with a plain wedgie (no inverting
blob) should trigger on the positive edge, but maybe that's too
simple.

H&H 2nd edition P508/509 states that a JK triggers on the negative
edge, and an edge-triggered FF on the positive edge. But it is not
clear whether that refers to the implementation as shown, or to thje
bare logic symbol. Most JK symbols do have the inverting blob on the
clock input.

P510 suggests that both dividers behave the same, which would imply
that the plain JK symbol triggers on the negative edge,

Yet P511 states that the shown ripple divider clocks on the negative
edge, the JK symbols do have the inverting blob, so the plain JK
symbol would clock on the positive edge.

My tenative conclusion is that a plain (no inverting blob) D FF clocks
on the positive edge. But on which edge clocks a plain JK FF?
Positive, neative, or worse: there is no standard?

Wouter van Ooijen

-- ------------------------------------
I think you maybe referring to a Master Slave JK FF.

the JK states are retained at the start of the Transition
and then used at the end of the transition of the clock and
thus sets the output then.
so this means in this case the output actually changes state
at the end of the "Said" state but using the values of the J&K at the
start !
where as the other type of JK simply uses the JK state at the start
of the transition.

I have an electronics quick text guide in PDF that covers this
area among others if you wish to see it.
it is very good and clear on explaining these things.

 

[normanstrong]

"Rich Grise" <null@example.net> wrote in message
news:2KX7d.4849$x65.1911@trnddc06...

On Saturday 02 October 2004 12:59 pm, Clarence did deign to grace us
with
the following:

If Bush fired everyone in government who made a mistake he would
have no

^^
one in the government at all,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Mistakes are inevitable in any large operation. It's what you do
after you discover the mistakes that makes all the difference.
Clearly, invading Iraq was a mistake. So far, I see no indication
that the Bush administration either recognizes that they made a
mistake, or is trying to rectify it.

Norm Strong

 

[Dave VanHorn]

Depends on what is "not much". In many of my cases the unit has to drop to
a few microamps when not in use. No switches or buttons. When operating
there would be about a 2-4mA budget allocation for the uC since that's
about what the circuitry it would replace consumes.

Seems reasonable.
The megas can operate from a 32kHz watch xtal, and have an XDIV register to
scale the clock by 1/N where N=1-129

You mean the eraser that is integrated into the top of a pencil? Wow, that
would indeed be small.

That's the one. I had to sacrifice efficiency to meet cost and size, but
hey, you have to compromise somewhere, and wall-plug power is essentially
free.

Does the circuit detect whether the paper rolls over or under? Just
kidding....

Actually, yes.
If you put the paper roll on the wrong way (over) then the print density
test won't pass since you'd be trying to print on the wrong side of the
paper.

Anyone with cats will tell you, that over is definitely the wrong way.

--
KC6ETE Dave's Engineering Page, www.dvanhorn.org
Microcontroller Consultant, specializing in Atmel AVR

 

[Rich Grise]

On Monday 04 October 2004 05:16 am, Fred Bartoli did deign to grace us with
the following:

"John Woodgate" <jmw@jmwa.demon.contraspam.yuk> a écrit dans le message de
news:KST3CDC5sQYBFw+g@jmwa.demon.co.uk...
I read in sci.electronics.design that Fred Bartoli <fred._canxxxel_this_
bartoli@RemoveThatAlso_free.fr_AndThisToo> wrote (in <41606ebd$0$24003$6
36a15ce@news.free.fr&gt about 'Marketing blurb - bullsh*t baffles etc',
on Sun, 3 Oct 2004:

Autoadaptative nheuristron )

Somewhat depressing...

Is the 'n' a typo or do we have to look for 'nheuris-' as a fun word in
some arcane language?

No typo. It was an attempt at mixing heuristic and the neuro-blah words
into a maybe obscur (and maybe 50's fashioned) technoword.

Simple neuristron or heuristron would have been too easily decipherable.
I find that the "nh" very improbable cooccurence and the confusion it sows
between "neuro" and "heuris" add a nice enigmatic touch.

But if it had been written in Vietnamese, there wouldn't have been anything
much special about it at all. ;-)

 

[John Woodgate]

I read in sci.electronics.design that Jonathan Kirwan
<jkirwan@easystreet.com> wrote (in <p583m0prhp9csl3a3ch5h3f7l37i37gq73@4
ax.com&gt about '48 Nobel Laureates Endorse Kerry', on Mon, 4 Oct 2004:

The group that is roughly the diametric opposite of the one that
bolsters the accumulation of wealth and power into fewer hands, I
suppose. This is neither Democrat or Republican, in the capital-letter
(proper noun) sense.

Sounds awfully like Old Labour, in British politics, to me. Does he
believe in:

"To secure for all the workers by hand or by brain the full fruits of
their industry and the most equitable distribution thereof that may be
possible upon the basis of the common ownership of the means of
production, distribution and exchange, and the best obtainable system of
popular administration and control of each industry of service."

..... the notorious wording of the former Clause 4 of the Labour Party
Constitution. For 'common' read 'Government'. For 'popular' read
'Government'.

It doesn't say that now. Even these democratic socialists have found
that governments are VERY BAD at running businesses. One might infer
that businessmen are VERY BAD at running governments, and there is some
truth in that, at least in a democracy.
--
Regards, John Woodgate, OOO - Own Opinions Only.
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

 

[John Woodgate]

I read in sci.electronics.design that Spehro Pefhany <speffSNIP@interlog
DOTyou.knowwhat> wrote (in <vk83m050pc7drdi4jtc3a1m90nenj04l19@4ax.com&gt
about 'The Rational Mind of Fred Bloggs', on Mon, 4 Oct 2004:

http://www.bartleby.com/65/co/conspira.html
Seems a bit more than talking about the weather is required...

I believe that is not so in English law, or at least it is arguable that
planning a legal act is still a conspiracy.

But anyways:-

Etymology: Middle English, from Middle French conspirer, from Latin
conspirare to be in harmony, conspire, from com- + spirare to breathe

I visualize people huddled together whispering to each other. According
to the OED (not a legal definition), it's possible for one person to
conspire.

Well, one person can aspire, expire, perspire, respire and suspire, so
why exclude conspire.
--
Regards, John Woodgate, OOO - Own Opinions Only.
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk