Driver to drive?

[Ken Smith]

In article <87f9f27c.0410060255.4fa14a4e@posting.google.com>,
Marco <billyliar@libero.it> wrote:

Hi to all,
in my research activities I need a suitable VCO working in the
frequency range between 1MHz - 200MHz.
Please, let me know what IC can be useful to this purpose.
Thank you in advance.
Marco.

A VCO that covers that range is likely to be very-very noisy and very
complex. Can you live with that? If not you may want to consider band
switching.

Circuits can be made to cover this wide of a range, continuously, but they
are not simple nor linear. You have to sort of "band fade" between the
components that work at the bottom end and those that work at higher
frequencies and then again to another set and so on until you get to the
top. Unless you really really need to do it, this is a method you should
avoid like the plague.



--
--
kensmith@rahul.net forging knowledge

 

[George W]

"Joerg" <notthisjoergsch@removethispacbell.net> wrote in message
news:_KV8d.6045$JG2.3231@newssvr14.news.prodigy.com...

<snip>

You don't have to do this with a DSP. It can also be done with a
detector. The trick is to provide a narrow enough filter to avoid false
triggers and there are many ways to do that. Directly at the RF level it
may be cumbersome because large resonators are bulky and expensive. I'd
probably down-convert to something reasonable, maybe in the 50MHz range
where you can achieve enough precision but can obtain or build filters
at reasonable expense. This has been a long time ago but TV sets contain
IF filters of about 5MHz bandwidth and because they are made by the
gazillion should be cheap. They are also carefully designed for a nice
flat group delay.

Of course the filter will shallow the slope of the onset. The lower the
bandwidth the more difficult it will be to obtain 50nsec precision. You
would need to accomodate for variances in RF signal strength which can
be done by registering the RF level once the detector has ramped up and
then adjust the measured slope position to compensate if the RF level
changed because of obstacles or whatever may be in the path.

I don't know the application but if it is short range and you can use an
ISM band with plenty of bandwidth there should not be too many issues
with interference.

That filter is gonna be tough to do in the tens of MHz without a DSP, I
think. I'm dealing with channels about 15 kHz center-to-center (only about
5 kHz guard bands) at 200 MHz. That's a pretty small percentage bandwidth,
and the filter has to have good group delay too. The receiver also has to
be able to tune to all the channels. So putting a detector up front and a
good filter in front of that, might be as difficult as using a DSP that does
both jobs. Will have to look hard at all the alternatives.

Prices have come down big time, mostly because of the advent of DSL and
cable modems. Check out TI. Yes, these DSP can do all the math you want
to perform afterwards on the same chip that does the detection. Also,
you may be able to pull routines from a library if the manufacturer
provides them. But in this application a DSP may still be overkill.

Thanks for the pointer, I would not have thought to look at components for
cable modems. The DSP I used a few years ago was an Analog Devices part,
but I am not up to date on what has become available since then.

Vielen dank

George

 

[Mike Monett]

John Larkin wrote:

I really don't understand your objection.

No, you don't. But instead of lecturing me, think about it for a
while.

John

That's what's so baffling. I know that you know the topic very well. It's
part of your company's business.

The examples listed showed different kinds of linearization. In most cases
the improvement could not be obtained with the addition of a simple
resistor or other component. It had to be done in software. None of the
problems you mentioned occurred.

Can you provide an example url that demonstrates the kind of problems you
are thinking of?

Mike Monett

 

[Ken Smith]

In article <ck28l905af@drn.newsguy.com>,
Winfield Hill <Winfield_member@newsguy.com> wrote:

Ken Smith wrote...

The J201 didn't work for us. We ended up talking Interfet into
producing (selecting) fets for us that are tuned to our need.

You found the J201 family different from the 2N4867 family?
Or did the low-noise screening issue get you? Please provide
more detail. Thanks!

It was both the screening and the shielding by the metal can that were at
issue. The J201's varied over a much wider range and had a lot more
finger effect.

Finger effect: The circuit works if I put my finger right here.

Consider the issues you have posed. Pass 250mA capability with
a low voltage drop (a volt more or less, which could be inside
the feedback loop),

Actually I can stand a bit more drop. I've got a 15V nominal supply and
could live with the 0.25A only working up to the swing of about 10V if I
had to.

but handle 600V peak. This implies the node
will experience a power level of greater than 0.25 * 600V = 150W
at the beginning of an event, which is clearly too much for small
components unless the time duration is very short, on the order
of milliseconds. This is a simple matter of energy-rate (power)
and thermal mass in the component withstanding the fault voltage.

"Simple matter" may be mis-stating it a bit. 150W spikes are not a simple
matter to deal with in most components. The idea isn't complex but
finding components can be. If I use a power MOSFET, I really need to
monitor the voltage in some way so that an attempt to pull the output
outside the rails causes the MOSFET to be shut off.

Your job is to find a way to protect the 600V PTC component, and
to prevent it from overheating as it handles an event. Fuzetec
is not specific: their FHR series says, "keep the time short."

The FHR ones are speced in what looks wrong to me. They are not rated to
interrupt at 600V. They are not rated to withstand 600V for very long.
This implies to me that 600V can never be applied to them. I'm sure its
some issue to do with one of the "standards" that causes the specs to be
weird.

Apparently this business of specifying thermal mass isn't easy, so
manufacturers give you a bit of data.

I've noticed that many makers don't really give you much data. They run a
few "typical" cases and document them.

--
--
kensmith@rahul.net forging knowledge

 

[Ken Smith]

In article <10m931vt1hbfd6@corp.supernews.com>,
Tim Wescott <tim@wescottnospamdesign.com> wrote:
[...]

So let's get this straight -- every component you design in goes
obsolete. So you're asking each of _us_ to recommend a favorite
solution, and the "winner" gets to see the components for his/her
favorite solution go obsolete?

Not everything goes obsolete. I designed in a nice Lithium battery in a
AA package.
(Purchasing guy enters room)

Ooops, I've got to go.

--
--
kensmith@rahul.net forging knowledge

 

[John S. Dyson]

In article <ck28ue05so@drn.newsguy.com>,
Winfield Hill <Winfield_member@newsguy.com> writes:

Tom Seim wrote...

Strange, I thought you were in favor of John Kerry. Curious.

No doubt, like me, he's simply strongly against G. W. Bush.

Scary thing is that many people who are against GWB have apparently

let the hatred damage their sanity (or at least, their credibility
about being rational.)

It is CERTAINLY okay to be against GWB (as I was simply against
Clinton because of his lack of personal self control, therefore
damaging his credibility.) I never developed a hatred or even
a serious dislike, even though my opinion was more an emotional
disgust about Clinton's 'dirty' behavior in/around the oval office,
during office hours, with an obviously immature woman/child. I
certainly wouldn't want to see Clinton impeached because of his
irresponsibility in that matter, but rather dishonesty and effective
perjury as a court officer...

Just as my dislike of Clinton for President wasn't absolute and
certainly not hate filled, my own support for the only credible
candidate for President, George W Bush, isn't unconditional. My
lack of support for Kerry is IN NO WAY hate filled, but rather
based upon Kerry's ongoing poor judgement and even poor attendance
in his primary occupation. Frankly, I am not happy with some of GWBs
decisions, but the reason for my unhappiness is based upon faulty
intelligence information (mostly resulting from a weakened intelligence
community, mostly from the involvement of the overly pacifist and
unrealistic leftists, where Kerry would be a typical offender.)

I certainly DO NOT hate Kerry, and wouldn't even be seriously
worried about him as Senator, Mayor, Governor or Dog Catcher.
When looking at (and evaluating) the current candidates for
president, with very rational and dispassionate (sp) evaluation,
then GWB is the only reasonable choice coming from the facts and
unemotional decision making. Geesh, if I had believed (incorrectly,
however), that Al Gore should really be president right now, and had
an emotional disturbance that kept me from seeing reality (perhaps
being misled by Democrat spin) that Bush really did win by a very
small margin of votes in just the right place, then I could possibly
imagine being irritated by the Bush presidency. However, the hatred that
is obviously manifested by obviously mentally ill, politically active
people does show that yet another category of misfits or dysfunctional
people (neurotic or even psychotic people) are constituents of the
Dem party.

John

 

[Rich.Andrews]

"Boki" <bokiteam@ms21.hinet.net> wrote in news:ck2c6m$4sp$1
@netnews.hinet.net:

short pin.8 pin.4 work...

yup. that will do it every time.

r



--
Nothing beats the bandwidth of a station wagon filled with DLT tapes.

 

[budgie]

On Wed, 06 Oct 2004 09:25:35 GMT, "Clarence" <no@No.com> wrote:

"budgie" <me@privacy.net> wrote in message
news:a9d7m0tadjgl44bd4hnlb94vasvktsbp9q@4ax.com...
On Wed, 06 Oct 2004 05:22:07 GMT, "Clarence" <no@No.com> wrote:


"budgie" <me@privacy.net> wrote in message
news:sjq6m0hj7mriscpa39tboc8ibhibf8k1d3@4ax.com...
On 5 Oct 2004 05:19:04 -0700, johankarremans@hotmail.com (Johan Karremans)
wrote:

Can somebody tell me what the substitute ic is from ic 74LS173.

74LS173 in stock at my local supplier here in Perth Oz at $A0.99. Why do
you
think you need a substitute - if I can get them down the corner, I'd be
surprised
if they aren't available almost everywhere.

http://www.jdr.com/interact/item.asp?itemno=GR-74LSG2

$0.39 ea

which would be more than I pay ($A0.99) each by the time I get them into
Australia.

$0.39 USD is about $A0.56 However, are you only going to buy one?
Sorry, I never buy less than a tube (abt 25) So I wouldn't even consider one.

If I were to buy a tube of every type I use here in proto work, there'd be no
room left for me. I buy in whatever quantity seems appropraite for the device
in question, based on the likelihood of not getting stuck with yet another 24
white elephants. Besides, my 25-up price is the same.

And I have no idea whether the O/P wanted one or had a production batch
requirement. But certainly MOQ and shipping costs ex US (or most other places)
make the local one-off price reasonably attractive.

 

[John Woodgate]

I read in sci.electronics.design that Travis Hayes <tmh-
SPAMISSPICEDHAM@nerdshack.com> wrote (in <2sjstiF1ldh5hU1@uni-
berlin.de&gt about 'Is this antenna article serious?', on Wed, 6 Oct
2004:

You are thinking of "Gold at the Starbow's End", by Fredrick Pohl.
The process described was called "engodelization", I suppose after
Godel. The back-of-envelope version of this concept is to represent
each character as a prime number, then send back the product of the
primes. The receiver has to factor the received data to recover the
primes used. For example, if I sent the number 2040, you would factor
it to see that I sent 2 2 2 3 5 17, and then figure out what I meant
with those 6 numbers.

An interesting prediction, because that is something like public key
encryption. But factoring 6 digit numbers doesn't take much computer
power, it's the 128 digit ones that take a few minutes. (;-)

Of course, the resulting set of prime numbers represents a very powerful
sub-code, because, for example, the three occurrences of '2' don't have
to mean the same thing. One of the primes might well determine the order
in which the other five are arranged.
--
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

 

[Kevin Aylward]

James Cheung wrote:

Hello dear electronics fans,

I am studying transimpedance amplifier (TIA) circuit for photodiodes
with bit rate approaching 2Gbps.

So far I have seen in the literature,
Op Amp like TIA IC are used to amplifier the current signals to
voltage
signals.

No chance. This is way to high for op-amp based designs on the front
end.

Their analysis are based upon such Op Amp configuration.

1. Is there any fundamental advantage of using IC to amplifier?

Generally its simpler.

The
downside is that the bit rate and sensitivity is controlled by the IC.

Yes.

2. I am thinking of using BJT in 2 stages to amplifier. Is there any
good literature on this topic? I expect those BJT's for RF use (e.g.
7GHz bandwidth) can extend the bit rate beyond 2Gbps.

This is a non trivial task. I would have a look at fets, especially
gasfets.

One technique is simple to use a fet source buffer, and let the signal
roll off due to the bias resister and capacitance of the diode/fet, then
amplifier the low impedance output and HP filter it to recover a flat
response.



Kevin Aylward
salesEXTRACT@anasoft.co.uk
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

 

[Rolavine]

From: bokiteam@ms21.hinet.net (boki)

I check it works or not by its TX and RX pins. If I short
pin.2(TX) and pin.3(RX) then I can see the RX signal received the
signal I send from TX( or say, there is a signal send out from TX), if
I open TX and RX, the TX is no output signal anymore....@@ (program is
still sending signal)

If you short pins 2 and 3, then can you see the signal on the scope? If the
signal path is looping correctly through the computer and you can't see the
signal on the scope, youf problem may be scope triggering. A lot of people
don't know how to use a scope.

On more possible though, some of the charge pump type RS232 drivers have a
power saving mode that may be switched on if they don't see a load on the
transmit line? The scope may not be enough of a load to enable the higher
voltage driver, I think I've worked with some of these. You could test this by
putting a 1k ohm resistor between the transmit pin and ground and then using
the scope on it, or by consulting the data sheet of the driver IC.

Rocky

 

[Paul Burke]

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

I think you misunderstand my question. I want to know the
interpretation of the symbol. A square with a wedgie in it, with a
line going into the wedgie. Is that supposed to denote a
positive-edge-triggered thingie or a negative-edge-triggered thingie?
Or does the answer depend on the kind of thingie, D or JK FF? Or dies
it also depend on old-style versus IEEE symbol?

I don't think we did. The answer is, yes, if the wedge has a circle,
things probably happen at the falling edge. But if it hasn't got a
circle, things may still happen at the falling edge- not as far as I
recall seeing in a datasheet, but certainly in schematics of circuits.
The designer hasn't bothered, or has forgotten, to put the circle in.

IEEE symbols, which were an attempt to sort all this out, never really
took off, at least in the circles I move in. Designed around the
computer and plotter capabilities of the 70s, and profoundly anti-
intuitive, they make most symbols look like a Dalek with a tram
pantograph, and the decoding of the ciphers within is a study in itself.
And in any case, circuit components rapidly got beyond the kind of thing
they can describe.

Paul Burke

 

[Rolavine]

From: bokiteam@ms21.hinet.net (boki)

Dear All,
I have a portable device that has a 5-V to 0-V pulse output(
without constant frequency, but below 10Hz range), I want this signal
could be detected by PC RS232 port.

I try many connection type, the RS232 still can't detect the device
pulse.

Do we have to design RS-232 handshake circuits with device output to
make the RS232 detect that pulse?

* I don't have to transmit data,just only pulse detection.

Is that baud rate problem? or ? @@

Your not going to get the uart to give you an serial input for that signal, but

if you check the internal registers you can prob. detect it as a framing or
overun erorr. I think you can also read the direct state of the receive pin
via registers, I haven't done that for about 4 years so I don't remember. I
know you can read the input of some of the other RS232 lines via these
registers.

Thank you very much for your help.

Best regards
Boki.

 

[Robert Baer]

Marlboro wrote:

Mark <Mark@> wrote in message news:<lgk7m01p79gglrit0i7o0d07aq7esng7ol@4ax.com>...

Sound like you have hard time, it happens to me once a while...
here some of my thoughs from what you posted


I am more of a digital man than an analogue one. I need help please,
the set-up is as follows, I have an serial ADC output feeding an
Op-Amp feeding a 12V 40W Amplifier IC. The amp works in Class H to get
the output. The trouble is I am getting HF oscillations at high volume
levels (in rythem with the signal). You can see them on a scope, I


Look like its your power supply issue, its not trong enough to power
the final Amp, since it works fine in low volume?


have a common ground plane for all of the circuit, I have thick tracks
going to the high power parts, I have a choke in line with the 7805
regulator for the logic. The Op-Amp is powered from the 5V logic line
(I only need 2V swing to drive output amplifier fully. I have tried
putting a heavy choke to the power supply input of the power amplifier
section, which did not help, I have tried putting a cap across the
input to the output amp, this helped a lot (with some decrease in
treble) but did not cure it completely, I have tried doubling the
values of the two caps used for the class H operation, did nothing,
tried doubling the power supply reservoir capacitor, this did nothing,
tried isolating the OP-Amp power supply with a diode and a smoothing
cap, this did nothing, also tried adding a choke to the opamp supply,
again nothing.


Have you try to power the opam & ADC from a COMPLETE separated supply
from the big Amp? This may give a clue of what going on. Another
though is that you may have ground/power plans/tracks improper design.
Review your PCB and check that if any output may have the "return
current" to your sensitive components such as opam & ADC...Try to
isolate the GND plan/or temporary re-route your tracks with wires if
necessary,


The thing that is throwing me is my Tektronics scope. When I put a
heavy connection wire between the Ground input of the scope and the
Ground of the circuit under test, and I put my probe onto the Ground
of the circuit under test (even at the same point where the heavy
ground wire is, I can see the HF oscillations on the screen. Trouble
is this is throwing me off the scent as to where the Oscillations are
coming from.

Can anybody give me some pointers please, I am starting to think that
I should have not taken on an Analogue design. Any web sites giving


Hmmm, dont give it up easy cauze you'll learn a great deal when you
beat the beast. Frankly I do not work with audio but video, and its
very sensitive to noise, osc,...specially when it is
hi-res/progressive scan. Most of the time
I found its the ADC/DAC issues, these chips need to be take care alot:
clean
power source, clean clock, good PCB design with proper GND/POWER
plans. The next important is always the power supply...


hints and tips on this subject would also be appreciated.


Anyway, it's hard to tell cauze can't see your circuit & PCB, but
believe there
always an answer to that, maybe other folks have better ideas.

cheers,

Regarding the power supply, adding an inductor to a 7805 may make
things worse.
Use two 0.22uF mylar caps, one right at the input and the other right
at the output of the 7805.
Then try disconnecting that power amp (the class H) to see if things
are OK.

 

[Boki]

"Rolavine" <rolavine@aol.com> źśźgŠóślĽóˇsťD:20041007030334.11452.00004064@mb-m23.aol.com...

From: bokiteam@ms21.hinet.net (boki)

I check it works or not by its TX and RX pins. If I short
pin.2(TX) and pin.3(RX) then I can see the RX signal received the
signal I send from TX( or say, there is a signal send out from TX), if
I open TX and RX, the TX is no output signal anymore....@@ (program is
still sending signal)


If you short pins 2 and 3, then can you see the signal on the scope? If
the
signal path is looping correctly through the computer and you can't see
the
signal on the scope, youf problem may be scope triggering. A lot of people
don't know how to use a scope.

On more possible though, some of the charge pump type RS232 drivers have a
power saving mode that may be switched on if they don't see a load on the
transmit line? The scope may not be enough of a load to enable the higher
voltage driver, I think I've worked with some of these. You could test
this by
putting a 1k ohm resistor between the transmit pin and ground and then
using
the scope on it, or by consulting the data sheet of the driver IC.

Rocky

Dear Rocky,
Ą@Ą@Ą@Ą@Ą@Sorry to ignore your first section, and I design a special
homework for you, that is to short RS232 pin.2 and pin.3 and check the scope
result. ( Assume Tx is sending signals )

I think this problem is " No short, no send..." even test Tx by scope
directly....

Best regards,
Boki.

 

[Boki]

§Ú¤ŁŞžšD§AŚbÁżÔŁ¤p

"Rolavine" <rolavine@aol.com> źśźgŠóślĽóˇsťD:20041007031415.11452.00004065@mb-m23.aol.com...

From: bokiteam@ms21.hinet.net (boki)

Dear All,
I have a portable device that has a 5-V to 0-V pulse output(
without constant frequency, but below 10Hz range), I want this signal
could be detected by PC RS232 port.

I try many connection type, the RS232 still can't detect the device
pulse.

Do we have to design RS-232 handshake circuits with device output to
make the RS232 detect that pulse?

* I don't have to transmit data,just only pulse detection.

Is that baud rate problem? or ? @@

Your not going to get the uart to give you an serial input for that
signal, but
if you check the internal registers you can prob. detect it as a framing
or
overun erorr. I think you can also read the direct state of the receive
pin
via registers, I haven't done that for about 4 years so I don't remember.
I
know you can read the input of some of the other RS232 lines via these
registers.

Thank you very much for your help.

Best regards
Boki.

 

[Boki]

Why @@?

pin.8 CTS
pin.4 DTR

?



"Rich.Andrews" <spmaway@ylhoo.com> źśźgŠóślĽóˇsťD:Xns957AF27B09BC4mc2500183316chgoill@10.232.1.1...

"Boki" <bokiteam@ms21.hinet.net> wrote in news:ck2c6m$4sp$1
@netnews.hinet.net:

short pin.8 pin.4 work...




yup. that will do it every time.

r



--
Nothing beats the bandwidth of a station wagon filled with DLT tapes.

 

[Mark]

On Thu, 07 Oct 2004 00:41:07 -0700, Robert Baer
<robertbaer@access4less.net> wrote:

Marlboro wrote:

Mark <Mark@> wrote in message news:<lgk7m01p79gglrit0i7o0d07aq7esng7ol@4ax.com>...

Sound like you have hard time, it happens to me once a while...
here some of my thoughs from what you posted


I am more of a digital man than an analogue one. I need help please,
the set-up is as follows, I have an serial ADC output feeding an
Op-Amp feeding a 12V 40W Amplifier IC. The amp works in Class H to get
the output. The trouble is I am getting HF oscillations at high volume
levels (in rythem with the signal). You can see them on a scope, I


Look like its your power supply issue, its not trong enough to power
the final Amp, since it works fine in low volume?


have a common ground plane for all of the circuit, I have thick tracks
going to the high power parts, I have a choke in line with the 7805
regulator for the logic. The Op-Amp is powered from the 5V logic line
(I only need 2V swing to drive output amplifier fully. I have tried
putting a heavy choke to the power supply input of the power amplifier
section, which did not help, I have tried putting a cap across the
input to the output amp, this helped a lot (with some decrease in
treble) but did not cure it completely, I have tried doubling the
values of the two caps used for the class H operation, did nothing,
tried doubling the power supply reservoir capacitor, this did nothing,
tried isolating the OP-Amp power supply with a diode and a smoothing
cap, this did nothing, also tried adding a choke to the opamp supply,
again nothing.


Have you try to power the opam & ADC from a COMPLETE separated supply
from the big Amp? This may give a clue of what going on. Another
though is that you may have ground/power plans/tracks improper design.
Review your PCB and check that if any output may have the "return
current" to your sensitive components such as opam & ADC...Try to
isolate the GND plan/or temporary re-route your tracks with wires if
necessary,


The thing that is throwing me is my Tektronics scope. When I put a
heavy connection wire between the Ground input of the scope and the
Ground of the circuit under test, and I put my probe onto the Ground
of the circuit under test (even at the same point where the heavy
ground wire is, I can see the HF oscillations on the screen. Trouble
is this is throwing me off the scent as to where the Oscillations are
coming from.

Can anybody give me some pointers please, I am starting to think that
I should have not taken on an Analogue design. Any web sites giving


Hmmm, dont give it up easy cauze you'll learn a great deal when you
beat the beast. Frankly I do not work with audio but video, and its
very sensitive to noise, osc,...specially when it is
hi-res/progressive scan. Most of the time
I found its the ADC/DAC issues, these chips need to be take care alot:
clean
power source, clean clock, good PCB design with proper GND/POWER
plans. The next important is always the power supply...


hints and tips on this subject would also be appreciated.


Anyway, it's hard to tell cauze can't see your circuit & PCB, but
believe there
always an answer to that, maybe other folks have better ideas.

cheers,


Regarding the power supply, adding an inductor to a 7805 may make
things worse.
Use two 0.22uF mylar caps, one right at the input and the other right
at the output of the 7805.
Then try disconnecting that power amp (the class H) to see if things
are OK.

I have the standard 104 caps right on the legs of the 7805, will try
linking out the inductor. Just to answer a previous post, yep I did
try powering the two sections with two completely different power
supplies, yes you quessed it, no joy!

 

[Norm Dresner]

Mark wrote:

I am more of a digital man than an analogue one. I need help please,
the set-up is as follows, I have an serial ADC output feeding an
Op-Amp feeding a 12V 40W Amplifier IC. The amp works in Class H to get
the output. The trouble is I am getting HF oscillations at high volume
levels (in rythem with the signal). You can see them on a scope, I
have a common ground plane for all of the circuit, I have thick tracks
going to the high power parts, I have a choke in line with the 7805
regulator for the logic. The Op-Amp is powered from the 5V logic line
(I only need 2V swing to drive output amplifier fully. I have tried
putting a heavy choke to the power supply input of the power amplifier
section, which did not help, I have tried putting a cap across the
input to the output amp, this helped a lot (with some decrease in
treble) but did not cure it completely, I have tried doubling the
values of the two caps used for the class H operation, did nothing,
tried doubling the power supply reservoir capacitor, this did nothing,
tried isolating the OP-Amp power supply with a diode and a smoothing
cap, this did nothing, also tried adding a choke to the opamp supply,
again nothing.

The thing that is throwing me is my Tektronics scope. When I put a
heavy connection wire between the Ground input of the scope and the
Ground of the circuit under test, and I put my probe onto the Ground
of the circuit under test (even at the same point where the heavy
ground wire is, I can see the HF oscillations on the screen. Trouble
is this is throwing me off the scent as to where the Oscillations are
coming from.

Can anybody give me some pointers please, I am starting to think that
I should have not taken on an Analogue design. Any web sites giving
hints and tips on this subject would also be appreciated.

I didn't see in all of this a characterization of the Op Amp's load. Is it
capacitive? Inductive? What OpAmp is it?

Do you have any overall HF feedback around the Op Amp? Depending on the
frequencies you're interested in (I'm assuming something near the audio
range) 15-30 pF shouldn't change the frequency response too much and would
be diagnostic.

Because you're a "digital man" you immediately think that output problems
like HF oscillation are caused by power supply issues but there are analog
circuits that delight in oscillating. In particular a complementary pair at
the output of an analog amp is a good example, especially if they're MOSFET.

Norm

 

[Winfield Hill]

John S. Dyson wrote...

Winfield Hill writes:
Tom Seim wrote...

Strange, I thought you were in favor of John Kerry. Curious.

No doubt, like me, he's simply strongly against G. W. Bush.

Scary thing is that many people who are against GWB have apparently
let the hatred damage their sanity (or at least, their credibility
about being rational.) [ snip ] ... and had an emotional
disturbance that kept me from seeing reality (perhaps being misled
by Democrat spin) that Bush really did win by a very small margin
of votes in just the right place, then I could possibly imagine
being irritated by the Bush presidency. However, the hatred that
is obviously manifested by obviously mentally ill, politically
active people does show that yet another category of misfits or
dysfunctional people (neurotic or even psychotic people) are
constituents of the Dem party.

The scary thing is the way you so readily let fly with completely
unwarranted assumptions and self-righteous criticisms about other
people about which you know very little indeed. And your ready
labeling of nearly half the population of the US with extreme
derogatory terms and language. That's scary.


--
Thanks,
- Win

(email: use hill_at_rowland-dotties-org for now)