开云体育

开云体育

A mark of a good moderator. Doesn't just tell supposed "off-topic" to
"shut up". But leads to forum where the "off-topic" is "on-topic".

Good Job!

Curtis

Dazzle Mom this coming Mother's Day season with flowers!

--- In Electronics_101@..., Stefan Trethan
<stefan_trethan@g...> wrote:

I have to disagree that a Spice simulators are not good at such
problems. Five minutes on one shows that the LT1028 will be unstable
in a transimpedace configuration, and that is before the capacitance
of the PMT is taken into account.

Simulators are only as good as the models and the circuit used. If
you don't include all your parasitics then they won't give the right
answer. If the design is at a high frequency then the capacitance of
each pad matters and so does the capacitance across a resistor.

that is exactly what i meant.
You need to be seriously skilled with it to take all that into account,
which i am not.

I say that it is definitely much harder to simulate unwanted effects

than

wanted effects
because wanted effects use components you know of, because you designed
and dimensioned them.
the unwanted effects use parasitic effects too, which you definitely
didn't design on purpose
and don't know that much off.
So to me it seems to be very much harder to simulate this problem

with the

oscillations than
for example the wanted amplification of the amplifier.

You say yourself you have the experience in this case from your work so
you can properly use
pspice for it. I would most likely just try to plug a capacitor on

and see

what happens in the
hardware.

Still i too strongly recommend getting that pspice because it is a

really

great tool.

ST

The problems, as shown by a quick simulation, are fundamental to the
opamp and the circuit configuration and do not require any details of
the parasitics. That may be required later, but I believe a quick
simulation is always worthwhile. It would have saved building a
circuit that doesn't work.

With regards to simulators, I would not recommend Pspice - it is
overpriced. There are plenty of others, including usable
demonstration versions, that cost from a small amount to nothing if
you just want some basic functionality.

Keith.

I have to disagree that a Spice simulators are not good at such
problems. Five minutes on one shows that the LT1028 will be unstable
in a transimpedace configuration, and that is before the capacitance
of the PMT is taken into account.

Simulators are only as good as the models and the circuit used. If
you don't include all your parasitics then they won't give the right
answer. If the design is at a high frequency then the capacitance of
each pad matters and so does the capacitance across a resistor.

that is exactly what i meant.
You need to be seriously skilled with it to take all that into account, which i am not.

I say that it is definitely much harder to simulate unwanted effects than wanted effects
because wanted effects use components you know of, because you designed and dimensioned them.
the unwanted effects use parasitic effects too, which you definitely didn't design on purpose
and don't know that much off.
So to me it seems to be very much harder to simulate this problem with the oscillations than
for example the wanted amplification of the amplifier.

You say yourself you have the experience in this case from your work so you can properly use
pspice for it. I would most likely just try to plug a capacitor on and see what happens in the
hardware.

Still i too strongly recommend getting that pspice because it is a really great tool.

ST

开云体育

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


You want longer wires? about 100 meteres? use a RS485 interface! lol!!
if that is too much for you, then follow the previows postings ;)


- --
Pablo E. Limon Garcia Viesca
ID 279930
plimon@...
al279930@...
pelgv@...

GPG-Key: 0x386906F1AD4C7FEA

- -----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GO d? S+:- a-- C++++>$ UL++++ P+>+++++
L+++>+++++>$ E--- W++ N++ w--- PS PE
Y+ PGP+++ t 5 X+ R- tv-- b+ D G e+ h*
r++ !y
- ------END GEEK CODE BLOCK------
www.geekcode.com
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)
Comment: Using GnuPG with Debian -

iD8DBQFAiUiqOGkG8a1Mf+oRAmgFAJ4kpQSJDPss+4tDzZIBcfLl07vpyACgl50p
KI0wYw4hblyYpjeNGjrNVyI=
=yBIq
-----END PGP SIGNATURE-----

--- In Electronics_101@..., Stefan Trethan
<stefan_trethan@g...> wrote:

On Fri, 23 Apr 2004 09:04:13 -0700 (PDT), k a <ktauyeung@y...> wrote:

The current source is from a PMT. What is a spice
simulator and what is it used for? The opamp I used is
a LT1028, very low noise. Thanks.

KT.

spice is a software for simulating electronic circuits.
it uses mathematical models for the parts.
You can either draw the schematic or input a netlist.


However, i would not suggest using it for the particular problem, as

in my

experience
this is exactly where the simulation differs from reality.
(may also be i am just not using the right models)

You should get one version of spice for other things, it is a great

help

in designing
circuits. You can experiment faster, at least with some circuits.

ST

I have to disagree that a Spice simulators are not good at such
problems. Five minutes on one shows that the LT1028 will be unstable
in a transimpedace configuration, and that is before the capacitance
of the PMT is taken into account.

Simulators are only as good as the models and the circuit used. If
you don't include all your parasitics then they won't give the right
answer. If the design is at a high frequency then the capacitance of
each pad matters and so does the capacitance across a resistor.

The phase margin of the LT1028 is not great for this sort of
application, but then most opamps need some form of compensation in
this application.

Unfortunately, the design of such amplifiers is quite complex (and one
area where I make my living). If you get a Spice simulator (I use
Simetrix from Catena software - a quick search should find it, and a
usable demo version) you could try different opamps to see if you can
find a suitable one. For a 1 Meg feedback, around 0.2pF would
stabilise it but that is not a practical capacitance. Other tricks
have to be used to get round the problem. If you can live with a
lower feedback resistance then the capacitor required to stabilise it
will be a more sensible value eg at 100k, 1pF would do. This does
reduce your signal to noise ratio and may require another gain stage
to get the signal you are after.

Keith.

As for inductors / transformers on these application notes - DON'T
try to do it yourself!

I disagree. I often wind my own inductors because what I want usually
doesn't exist. For power inductors I quite like the Micrometals E
cores becuase they have high energy storage without having to mess
about with gapping and are cheap. If you want gapped ferrites then
MMG Neosid do a good range.

i disagree too / agree with keith.
Buying the demo board is no way of learning how to do it.
it is a way of buying a ready-made unit, but then i can just go out and buy
the supply module and that would maybe be even cheaper.

To answer the issue of measuring the inductance, it is easy to do. I
often use resonance - put a known capacitor across it, a series
resistor and hook it up to a scope and signal generator. However, you
shouldn't need to measure it. Design it, wind it, use it.

Best approach if you have a scope and a generator.
You can test it close to the frequency you will use it.

If you use a small component tester you might want to make sure the measuring frequency
isn't so far off that the result is worthless. of course you can calculate the error.

ST

On Fri, 23 Apr 2004 15:34:15 -0000, upand_at_them <upand_at_them@...> wrote:

They make header posts that you can break off what you need, but I
have yet to see the female part built that way.

this is one of the messages where i don't know what this answer is referring to.
might be a good idea to add at least the one xentence containing the question.


there are break off female header receptors. there are also ones that can be cut apart,
but with these you loose the pin where you cut.
the breackaway version has no lost pin, but i have only seen them single-row so far.
if you need 2 row either use 2 parallel or use "cut apart" type.
from my experience the breakaway type has slightly worse contacts (metal only
on 2 sides).



What i find hard to get hold of are female headers for attaching cables, the ones
for flat cables which are self cutting are readily available but the type with individual
contacts which you crimp on / solder is difficult to get.
the contacts itself are not the problem but the plastic housing is hard to find.

ST

On Fri, 23 Apr 2004 09:04:13 -0700 (PDT), k a <ktauyeung@...> wrote:

The current source is from a PMT. What is a spice
simulator and what is it used for? The opamp I used is
a LT1028, very low noise. Thanks.

KT.

spice is a software for simulating electronic circuits.
it uses mathematical models for the parts.
You can either draw the schematic or input a netlist.


However, i would not suggest using it for the particular problem, as in my experience
this is exactly where the simulation differs from reality.
(may also be i am just not using the right models)

You should get one version of spice for other things, it is a great help in designing
circuits. You can experiment faster, at least with some circuits.

ST

--- In Electronics_101@..., "Bruce Carter" <brucec@m...>
wrote:

Probably. Give me a link to the data sheet and I will look it up
for you. Transformers are usually for isolation, if you are willing
to give up isolation you are probably all right.

As for inductors / transformers on these application notes - DON'T
try to do it yourself! The characteristics are usually so
intertwined that the manufacturer recommends vendor and part
number. Substitute and the circuit won't work. They usually sell
evaluation boards with the necessary parts - a very annoying, but
necessary investment for the hobbyist, because your chances of
getting samples of the oddball inductors, special capacitors,
schottky diodes, wierd transistors are almost zero - much less
making a PC board layout without parasitics that will ruin your
design.

I disagree. I often wind my own inductors because what I want usually
doesn't exist. For power inductors I quite like the Micrometals E
cores becuase they have high energy storage without having to mess
about with gapping and are cheap. If you want gapped ferrites then
MMG Neosid do a good range.

To answer the issue of measuring the inductance, it is easy to do. I
often use resonance - put a known capacitor across it, a series
resistor and hook it up to a scope and signal generator. However, you
shouldn't need to measure it. Design it, wind it, use it.

I am not quite sure of what you want to get out of the regulator. If
it is for car use and it runs off 12V, why regulate? Does the
equipment you want to use require a very precise voltage?

Keith.

The current source is from a PMT. What is a spice
simulator and what is it used for? The opamp I used is
a LT1028, very low noise. Thanks.

KT.

--- Keith <keith@...> wrote:

--- In Electronics_101@..., "ktauyeung"
<ktauyeung@y...>
wrote:

I am working on a high-gain current-to-voltage

amplifier. I need to

convert a very low pulsed current signal (100

microsec, <nanoamp)

into a voltage signal (>10 mV) that I can read

from the scope. I

tried the amplifier using a low-gain (0.2

V/microamp) setting, and it

works ok. When I tried it with a higher gain (2

V/microamp),

oscillations (~10 microsec, regular amplitude)

showed up on the scope

display. I'd like to fix that, but don't know how.

I need some help,

if you know how to fix it and what is the cause of

that, please let

me know. Thanks!

Is the current source a photodiode? If so, that is
probably the cause
of the oscillations. If you have access to a Spice
simulator you
could quickly check it. What opamp are you using?
2M ohms is NOT a
large resistor for such an amplifier. They can
easily be 10M or 100M.
If you use two stages you could be dogged by noise
problems. The
dominant noise component is usually the feedback
resistor. Reducing
this by a factor if 10 will only reduce the noise by
a factor of
sqrt(10) so you are roughly 3 times worse off for
signal to noise
ratio. Reduce it by 100 and you are 10 times worse
off. There is
also opamp current and voltage noise to consider.

Keith.

__________________________________
Do you Yahoo!?
Yahoo! Photos: High-quality 4x6 digital prints for 25?

开云体育

They make header posts that you can break off what you need, but I
have yet to see the female part built that way.

Probably. Give me a link to the data sheet and I will look it up
for you. Transformers are usually for isolation, if you are willing
to give up isolation you are probably all right.

As for inductors / transformers on these application notes - DON'T
try to do it yourself! The characteristics are usually so
intertwined that the manufacturer recommends vendor and part
number. Substitute and the circuit won't work. They usually sell
evaluation boards with the necessary parts - a very annoying, but
necessary investment for the hobbyist, because your chances of
getting samples of the oddball inductors, special capacitors,
schottky diodes, wierd transistors are almost zero - much less
making a PC board layout without parasitics that will ruin your
design.

How can you measure the inductance? Incase I choose to make my own
tranformer? The problem is I will need 2 (for current requirements) and
after the regulators the magnetics are the most expensive part of the
circuit by half.

You can get meters, looking like digital multimeters, that can measure this.
they are in about the same price range as digital multimeters.
the professional version is a "resistance, inductance, capacitance measuring bridge"
but a simple component meter is ok.
You can even get multimeters with inductance measuring range.

Another approach is with a scope and a ac signal.


But you can buy ready-made inductors for which the value is known.


What do you need the 12V for?
You should be able to use 5V components, or use a regulator only limiting, no stepup,
for motors (fans, maybe harddisk).


If you take apart a old PC power supply you will find a inductor which carries one winding
for 5V and one for 12V, maybe even one for 3,3V.
you can use it , using the 5V winding for primary and the 12V winding for secondary.


ST

I don't know about the LM2587 and would need more details of what you
are trying to do. If you are connecting to mains voltage then you

No, no mains at all, it is for in car use only.

need a transformer for safety/isolation. If you are converting from a
car battery or something like that then you can just use an inductor -
a transofrmer is not necessary. I do not know if the LM2587 is

Yes, it will be in car, the data sheet says that for the input voltage I
am looking at (10-15v) I have to use it in flyback mode, all the example
circuits have a transformer in them.

suitable for that configuration as I have never used it, but certainly
other switching regulator chips are.

How can you measure the inductance? Incase I choose to make my own
tranformer? The problem is I will need 2 (for current requirements) and
after the regulators the magnetics are the most expensive part of the
circuit by half.

--
Anthony Toft <toftat@...>

--- In Electronics_101@..., Anthony Toft <toftat@c...> wrote:

I am trying to desing a power supply, including a 12v flyback regulator
(it's for my incar mp3 player).

The regulator I have is a national LM2587, there are designs in the app
notes but all include a transformer. As I think I'll need more than one
regulator (for more current) I'd prefer not to. Is it possible to built
one with an inductor?



--
Anthony Toft <toftat@c...>

I don't know about the LM2587 and would need more details of what you
are trying to do. If you are connecting to mains voltage then you
need a transformer for safety/isolation. If you are converting from a
car battery or something like that then you can just use an inductor -
a transofrmer is not necessary. I do not know if the LM2587 is
suitable for that configuration as I have never used it, but certainly
other switching regulator chips are.

Keith.

--- In Electronics_101@..., JanRwl@A... wrote:

In a message dated 4/22/2004 11:02:17 PM Central Standard Time,
ksrao_iete@y... writes:
wanna discuss something very Seriously Teleportation.... yah.....

Did someone say "substance-abuse"???

naw but I did find the use of serious and teleport in the same
sentence some what oxymoronic...