What are these nodes?

[Jim Thompson]

What are these nodes?

http://tinyurl.com/bjxndpg

...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.

[Phil Hobbs]

On 01/26/2013 12:18 PM, Jim Thompson wrote:
> What are these nodes?
>
> http://tinyurl.com/bjxndpg
>
> ...Jim Thompson

AC analysis relies on linearity, so all signals are noninteracting.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

[Jim Thompson]

On Sat, 26 Jan 2013 12:53:37 -0500, Phil Hobbs
<pcdhSpamM...@electrooptical.net> wrote:

>On 01/26/2013 12:18 PM, Jim Thompson wrote:
>> What are these nodes?
>>
>> http://tinyurl.com/bjxndpg
>>
>> ...Jim Thompson
>
>AC analysis relies on linearity, so all signals are noninteracting.
>
>Cheers
>
>Phil Hobbs

AC analysis is indeed "small signal". That doesn't negate the fact
that the analysis finds a frequency misbehavior.

And I don't have a clue what your "non-interacting" means. Sounds
like BS to me >:-}

I'll run a transient analysis and prove it to you.

Sorry "prove" is the wrong word. Some on this group think their
opinion outweighs science.

[bitrex]

On 1/26/2013 12:18 PM, Jim Thompson wrote:
> What are these nodes?
>
> http://tinyurl.com/bjxndpg
>
> ...Jim Thompson
>

I lurk here mostly, but I don't understand the big deal over this
circuit. Isn't it just a poor man's TIA? Input current against R5
creates current in the output -Iin*R5/R4 so (neglecting the small
current contribution from Iin) Vo/Iin = -(R5*(R3+R4))/R4?

[John Larkin]

On Sat, 26 Jan 2013 10:18:52 -0700, Jim Thompson
<To-Email-Use-Th...@On-My-Web-Site.com> wrote:

>What are these nodes?
>
> http://tinyurl.com/bjxndpg
>
> ...Jim Thompson

Depends on what you mean by "AC Analysis". Most people are interested in the
overall transfer function of an amplifier, so the left side of C2 isn't ground,
it's the input.

The transfer function is OUT/INPUT, which isn't useful if you define the INPUT
node to be ground.


--

John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

[Jim Thompson]

Yes, exactly! A breath of fresh air from someone who knows what they
are doing, instead of BS'ing!

The big deal is that the "bootstrap" is not broadband. It is actually
a "poor man's gyrator", as I've demonstrated. I'm guessing that the
component values were chosen specifically to enhance behavior at
40KHz... that is, the sensor is likely a motion detector type.

So, as a plain vanilla amplifier, it's performance is terrible...
unless driven from a high impedance source... then it's pretty much
useless.

[John Larkin]

On Sat, 26 Jan 2013 13:11:11 -0500, bitrex <bit...@de.lete.earthlink.net> wrote:

James assumed that the signal source was a voltage, whose source impedance,
including the coupling cap C2, was far lower than the 220K bias resistor, so the
thing runs open-loop with a voltage gain of hundreds. The whole point of
bootstrapping was to increase the overall voltage gain. The feedback is just to
set the DC operating point, not to limit the gain or to make a TIA. As James
said, if AC feedback were a concern, add a bypass cap in the feedback path so
that it isn't.

The true poor man's TIA is a single resistor to ground.

[Jim Thompson]

On Sat, 26 Jan 2013 11:24:11 -0700, Jim Thompson
<To-Email-Use-Th...@On-My-Web-Site.com> wrote:

>On Sat, 26 Jan 2013 13:11:11 -0500, bitrex
><bit...@de.lete.earthlink.net> wrote:
>
>>On 1/26/2013 12:18 PM, Jim Thompson wrote:
>>> What are these nodes?
>>>
>>> http://tinyurl.com/bjxndpg
>>>
>>> ...Jim Thompson
>>>
>>
>>I lurk here mostly, but I don't understand the big deal over this
>>circuit. Isn't it just a poor man's TIA? Input current against R5
>>creates current in the output -Iin*R5/R4 so (neglecting the small
>>current contribution from Iin) Vo/Iin = -(R5*(R3+R4))/R4?
>
>Yes, exactly! A breath of fresh air from someone who knows what they
>are doing, instead of BS'ing!
>
>The big deal is that the "bootstrap" is not broadband. It is actually
>a "poor man's gyrator", as I've demonstrated. I'm guessing that the
>component values were chosen specifically to enhance behavior at
>40KHz... that is, the sensor is likely a motion detector type.
>
>So, as a plain vanilla amplifier, it's performance is terrible...
>unless driven from a high impedance source... then it's pretty much
>useless.
>
> ...Jim Thompson

Which is what Bitrex said, and I missed emphasizing, input "current".

As a voltage amplifier it's a differentiator.

[Phil Hobbs]

On 1/26/2013 1:05 PM, Jim Thompson wrote:
> On Sat, 26 Jan 2013 12:53:37 -0500, Phil Hobbs
> <pcdhSpamM...@electrooptical.net> wrote:
>
>> On 01/26/2013 12:18 PM, Jim Thompson wrote:
>>> What are these nodes?
>>>
>>> http://tinyurl.com/bjxndpg
>>>
>>> ...Jim Thompson
>>
>> AC analysis relies on linearity, so all signals are noninteracting.
>>
>> Cheers
>>
>> Phil Hobbs
>
> AC analysis is indeed "small signal". That doesn't negate the fact
> that the analysis finds a frequency misbehavior.
>
> And I don't have a clue what your "non-interacting" means. Sounds
> like BS to me >:-}
>
> I'll run a transient analysis and prove it to you.
>
> Sorry "prove" is the wrong word. Some on this group think their
> opinion outweighs science.
>
> ...Jim Thompson
>

I agree that running a simulation doesn't prove a whole lot by itself,
but then I don't have a dog in this fight.

"Non-interacting" means what it says--in a perfectly linear system, the
response to any combination of stimuli, applied wherever you like,
equals the sum of the responses to each stimulus alone.

I do build a whole lot of bootstraps, though.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203

Briarcliff Manor NY 10510 USA
+1 845 480 2058

[Jim Thompson]

On Sat, 26 Jan 2013 15:29:38 -0500, Phil Hobbs

<pcdhSpamM...@electrooptical.net> wrote:

>On 1/26/2013 1:05 PM, Jim Thompson wrote:
>> On Sat, 26 Jan 2013 12:53:37 -0500, Phil Hobbs
>> <pcdhSpamM...@electrooptical.net> wrote:
>>
>>> On 01/26/2013 12:18 PM, Jim Thompson wrote:
>>>> What are these nodes?
>>>>
>>>> http://tinyurl.com/bjxndpg
>>>>
>>>> ...Jim Thompson
>>>
>>> AC analysis relies on linearity, so all signals are noninteracting.
>>>
>>> Cheers
>>>
>>> Phil Hobbs
>>
>> AC analysis is indeed "small signal". That doesn't negate the fact
>> that the analysis finds a frequency misbehavior.
>>
>> And I don't have a clue what your "non-interacting" means. Sounds
>> like BS to me >:-}
>>
>> I'll run a transient analysis and prove it to you.
>>
>> Sorry "prove" is the wrong word. Some on this group think their
>> opinion outweighs science.
>>
>> ...Jim Thompson
>>
>I agree that running a simulation doesn't prove a whole lot by itself,
>but then I don't have a dog in this fight.
>
>"Non-interacting" means what it says--in a perfectly linear system, the
>response to any combination of stimuli, applied wherever you like,
>equals the sum of the responses to each stimulus alone.

PhD-ish for the superposition theorem ?:-)

>
>I do build a whole lot of bootstraps, though.

Whoopee Doo! I build thousands of cascoded mirrors.

>
>Cheers
>
>Phil Hobbs

[Phil Hobbs]

You really are pretty torqued about this, aren't you?

[Jim Thompson]

On Sat, 26 Jan 2013 15:29:38 -0500, Phil Hobbs

<pcdhSpamM...@electrooptical.net> wrote:

>On 1/26/2013 1:05 PM, Jim Thompson wrote:
>> On Sat, 26 Jan 2013 12:53:37 -0500, Phil Hobbs
>> <pcdhSpamM...@electrooptical.net> wrote:
>>
>>> On 01/26/2013 12:18 PM, Jim Thompson wrote:
>>>> What are these nodes?
>>>>
>>>> http://tinyurl.com/bjxndpg
>>>>
>>>> ...Jim Thompson
>>>
>>> AC analysis relies on linearity, so all signals are noninteracting.
>>>
>>> Cheers
>>>
>>> Phil Hobbs
>>
>> AC analysis is indeed "small signal". That doesn't negate the fact
>> that the analysis finds a frequency misbehavior.
>>
>> And I don't have a clue what your "non-interacting" means. Sounds
>> like BS to me >:-}
>>
>> I'll run a transient analysis and prove it to you.
>>
>> Sorry "prove" is the wrong word. Some on this group think their
>> opinion outweighs science.
>>
>> ...Jim Thompson
>>
>I agree that running a simulation doesn't prove a whole lot by itself,

Almost missed that jewel. Are you saying opinion outweighs
simulation?

>but then I don't have a dog in this fight.

Just your nose ?:-)

[snip]

[Jim Thompson]

On Sat, 26 Jan 2013 15:34:11 -0500, Phil Hobbs

Perhaps! I'm thinking of writing a book, "America's Worst
Engineering" and quoting from this public forum... citing names along
with the examples :-)

The "bootstrap", as presented, works only with a capacitive sensor and
is crap otherwise.

When I posted my misgivings, I was dumped on, rather than _anyone_
presenting a proper analysis.

[Phil Hobbs]

On 1/26/2013 3:34 PM, Jim Thompson wrote:
> On Sat, 26 Jan 2013 15:29:38 -0500, Phil Hobbs
> <pcdhSpamM...@electrooptical.net> wrote:
>
>> On 1/26/2013 1:05 PM, Jim Thompson wrote:
>>> On Sat, 26 Jan 2013 12:53:37 -0500, Phil Hobbs
>>> <pcdhSpamM...@electrooptical.net> wrote:
>>>
>>>> On 01/26/2013 12:18 PM, Jim Thompson wrote:
>>>>> What are these nodes?
>>>>>
>>>>> http://tinyurl.com/bjxndpg
>>>>>
>>>>> ...Jim Thompson
>>>>
>>>> AC analysis relies on linearity, so all signals are noninteracting.
>>>>
>>>> Cheers
>>>>
>>>> Phil Hobbs
>>>
>>> AC analysis is indeed "small signal". That doesn't negate the fact
>>> that the analysis finds a frequency misbehavior.
>>>
>>> And I don't have a clue what your "non-interacting" means. Sounds
>>> like BS to me >:-}
>>>
>>> I'll run a transient analysis and prove it to you.
>>>
>>> Sorry "prove" is the wrong word. Some on this group think their
>>> opinion outweighs science.
>>>
>>> ...Jim Thompson
>>>
>> I agree that running a simulation doesn't prove a whole lot by itself,
>
> Almost missed that jewel. Are you saying opinion outweighs
> simulation?

No, just that simulation isn't reality. You know perfectly well the
level of effort required to make a decent set of models for nontrivial
devices--it sure doesn't happen by accident. All those models have to
be validated by experiment (physical and numerical), and they only work
over a limited set of input parameters. For instance, zenering a BE
junction will reduce beta, but good luck getting that into a SPICE model.

>
>> but then I don't have a dog in this fight.
>
> Just your nose ?:-)

You know, Jim, the last time I checked, Usenet was a public forum. If
you want your tiffs to be private, you might consider conducting them
over the phone.

[Jim Thompson]

On Sat, 26 Jan 2013 15:44:46 -0500, Phil Hobbs
<pcdhSpamM...@electrooptical.net> wrote:

>On 1/26/2013 3:34 PM, Jim Thompson wrote:
>> On Sat, 26 Jan 2013 15:29:38 -0500, Phil Hobbs
>> <pcdhSpamM...@electrooptical.net> wrote:

[snip]

>>>>
>>> I agree that running a simulation doesn't prove a whole lot by itself,
>>
>> Almost missed that jewel. Are you saying opinion outweighs
>> simulation?
>
>No, just that simulation isn't reality. You know perfectly well the
>level of effort required to make a decent set of models for nontrivial
>devices--it sure doesn't happen by accident. All those models have to
>be validated by experiment (physical and numerical), and they only work
>over a limited set of input parameters. For instance, zenering a BE
>junction will reduce beta, but good luck getting that into a SPICE model.

Getting a bit off-track there aren't we?

Are you claiming that my simulation isn't reality? That there's
something in this simple circuit that is black magic, such that jerks
can claim a performance not demonstrated by a simulation?

I've just been vindicated by a Marshall Leach paper.

>
>>
>>> but then I don't have a dog in this fight.
>>
>> Just your nose ?:-)
>
>You know, Jim, the last time I checked, Usenet was a public forum. If
>you want your tiffs to be private, you might consider conducting them
>over the phone.
>
>Cheers
>
>Phil Hobbs

[amdx]

>>
>> You really are pretty torqued about this, aren't you?
>>
>> Cheers
>>
>> Phil Hobbs
>
> Perhaps! I'm thinking of writing a book, "America's Worst
> Engineering" and quoting from this public forum... citing names along
> with the examples :-)
>
> The "bootstrap", as presented, works only with a capacitive sensor and
> is crap otherwise.

Here's some, IMHO, relevant info the OP wrote, which seems to go agree
with your last line.


The input impedance doesn't matter--the input
source is a low-impedance highly capacitive transducer.
____________________________________________________________
It has nothing to do with gyrators, feedback, etc. Those all totally
miss the boat.
______________________________________________________________________
As far as the original problem which inspired this--which is entirely
irrelevant to the thread, and which was never going to use this ckt
anyhow--all I wanted was a spike out when the transducer got pinged.
____________________________________________________________________
That wasn't my assumption; I assumed a capacitive, low-impedance
source.
___________________________________________________________________
freq as spec'd in the drawings and all the results--40KHz. Source
impedance WAG'd @ 10nF, e.s.r. unknown (believed < 100r).

****************************************************************

I looked at about seven 40khz transducers they were all ~2000pf.
10nf seems high???
Mikek

[Jim Thompson]

On Sat, 26 Jan 2013 15:50:57 -0600, amdx <am...@knologynotthis.net>
wrote:

>
>>>
>>> You really are pretty torqued about this, aren't you?
>>>
>>> Cheers
>>>
>>> Phil Hobbs
>>
>> Perhaps! I'm thinking of writing a book, "America's Worst
>> Engineering" and quoting from this public forum... citing names along
>> with the examples :-)
>>
>> The "bootstrap", as presented, works only with a capacitive sensor and
>> is crap otherwise.
>
> Here's some, IMHO, relevant info the OP wrote, which seems to go agree
>with your last line.
>
>
> The input impedance doesn't matter--the input
>source is a low-impedance highly capacitive transducer.

I so noted that when flipper was pontificating.

>____________________________________________________________
>It has nothing to do with gyrators, feedback, etc. Those all totally
>miss the boat.

Except it doesn't. I've plainly shown that the "bootstrap" is a
gyrator-created inductance.

Then someone threw up another style of gyrator that wasn't as good.

>______________________________________________________________________
>As far as the original problem which inspired this--which is entirely
>irrelevant to the thread, and which was never going to use this ckt
>anyhow--all I wanted was a spike out when the transducer got pinged.

Fit's with my resonance" comments, doesn't it?

>____________________________________________________________________
>That wasn't my assumption; I assumed a capacitive, low-impedance
>source.

And I agreed. Capacitor with signal as a current source... since
vindicated by Leach's paper.

>___________________________________________________________________
>freq as spec'd in the drawings and all the results--40KHz. Source
>impedance WAG'd @ 10nF, e.s.r. unknown (believed < 100r).
>
>****************************************************************
>
>I looked at about seven 40khz transducers they were all ~2000pf.
>10nf seems high???
> Mikek

I guess. I'm not sure. I have some 30 year old ones in my parts bin
from a motion detector project.

[David Eather]

On Sun, 27 Jan 2013 03:18:52 +1000, Jim Thompson
<To-Email-Use-Th...@on-my-web-site.com> wrote:

> What are these nodes?
>
> http://tinyurl.com/bjxndpg
>
> ...Jim Thompson

Isn't the thevein equivalent of what those nodes are connected to a short
circuit?

[Jim Thompson]

On Sun, 27 Jan 2013 08:56:08 +1000, "David Eather" <eat...@tpg.com.au>
wrote:

I guess you might say that. What are they "shorted" TO?

In an AC analysis, all _FIXED_ voltage sources are presumed ZERO.

[David Eather]

On Sun, 27 Jan 2013 09:01:56 +1000, Jim Thompson

<To-Email-Use-Th...@on-my-web-site.com> wrote:

> On Sun, 27 Jan 2013 08:56:08 +1000, "David Eather" <eat...@tpg.com.au>
> wrote:
>
>> On Sun, 27 Jan 2013 03:18:52 +1000, Jim Thompson
>> <To-Email-Use-Th...@on-my-web-site.com> wrote:
>>
>>> What are these nodes?
>>>
>>> http://tinyurl.com/bjxndpg
>>>
>>> ...Jim Thompson
>>
>>
>> Isn't the thevein equivalent of what those nodes are connected to a
>> short
>> circuit?
>
> I guess you might say that. What are they "shorted" TO?
>
> In an AC analysis, all _FIXED_ voltage sources are presumed ZERO.
>
> ...Jim Thompson

If you stick thevein in then it is clearly shorted to gnd so it must be
equivalent to zero

[Jim Thompson]

On Sun, 27 Jan 2013 09:16:54 +1000, "David Eather" <eat...@tpg.com.au>

wrote:

>On Sun, 27 Jan 2013 09:01:56 +1000, Jim Thompson
><To-Email-Use-Th...@on-my-web-site.com> wrote:
>
>> On Sun, 27 Jan 2013 08:56:08 +1000, "David Eather" <eat...@tpg.com.au>
>> wrote:
>>
>>> On Sun, 27 Jan 2013 03:18:52 +1000, Jim Thompson
>>> <To-Email-Use-Th...@on-my-web-site.com> wrote:
>>>
>>>> What are these nodes?
>>>>
>>>> http://tinyurl.com/bjxndpg
>>>>
>>>> ...Jim Thompson
>>>
>>>
>>> Isn't the thevein equivalent of what those nodes are connected to a
>>> short
>>> circuit?
>>
>> I guess you might say that. What are they "shorted" TO?
>>
>> In an AC analysis, all _FIXED_ voltage sources are presumed ZERO.
>>
>> ...Jim Thompson
>
>If you stick thevein in then it is clearly shorted to gnd so it must be
>equivalent to zero

Only if the voltage source value goes to zero. A Thevenin equivalent
is a voltage in series with a resistor of a value determined by what
you get when you short the source.