fiddled filter design - Page 8

Do you have a question? Post it now! No Registration Necessary

Translate This Thread From English to

Threaded View
Re: fiddled filter design
On Friday, November 9, 2018 at 3:35:35 PM UTC+11, John Larkin wrote:
Quoted text here. Click to load it
ev
no
t
RM
as
 I
Quoted text here. Click to load it
  
Quoted text here. Click to load it
  
Quoted text here. Click to load it
  
Quoted text here. Click to load it
  
Quoted text here. Click to load it

There have been computer programs that have "evolved" circuits.  

http://citeseerx.ist.psu.edu/viewdoc/download?doi10%.1.1.109.6208&rep=r
ep1&type=pdf

Nobody seems to sell them.

Quoted text here. Click to load it
  
Quoted text here. Click to load it

Lots of stuff is way too intellectual for John Larkin.

Quoted text here. Click to load it
um. We need to make a good-enough filter from parts that we have in stock,  
ideally parts that are already on the board's BOM.

The need being to have stuff to sell that more or less works.

John Larkin has said that his average product development time is two weeks
, which doesn't leave a lot of time for exploring alternatives, finding an  
optimal solution, or even defining the problem to be solved clearly enough  
that you could demonstrate that a particular solution was optimal.

--  
Bill Sloman, Sydney


Re: fiddled filter design
Quoted text here. Click to load it

Do you have NuHertz?


Quoted text here. Click to load it

Bingo.

Actually, proof is probably a lot more boring than that.  You can  
interpolate the trajectory of poles between configurations in a very  
straightforward way.  The standard types all have poles lying on an ellipse  
enclosing the origin, giving two degrees of freedom (offset and  
eccentricity).

Well, for a given set of constraints (e.g., amplitude and phase flatnesses,  
sharpness of cutoff or stopband attenuation), there's probably a global  
minima which is slightly off an intermediate ellipse, and therein would lie  
the difficulty.

Tim

--  
Seven Transistor Labs, LLC
Electrical Engineering Consultation and Design
We've slightly trimmed the long signature. Click to see the full one.
Re: fiddled filter design
On Fri, 9 Nov 2018 18:29:27 -0600, "Tim Williams"

Quoted text here. Click to load it

We have their LC filter designer. The UI is clunky but it makes great
filters.


--  

John Larkin         Highland Technology, Inc
picosecond timing   precision measurement  

We've slightly trimmed the long signature. Click to see the full one.
Re: fiddled filter design

On 11/08/2018 05:42 PM, John Larkin wrote:
Quoted text here. Click to load it




Well... not really... the fundamental reason these filter came about was  
because they have closed form solutions of arbitrary order to satisfy a key  
specification. It was because there were no computers at the time

Today, its more down to laziness why other structures are not used.

Quoted text here. Click to load it

Actually, the Bessel filter was invented to solve the perfect delay, linear  
phase problem. The fact that it rolls off was a nuisance in its original  
construction. It was to synthesis of exp(-tau.s). It wasn't meant to be an  
amplitude filter at all.

With that in mind, the optimum 2nd order delay filter has a Q=2/pi, which is  
not the same as a Bessel order two. This number is found by doing a min/max  
analyses, which escapes me right now... I was 21 when I first came across  
that result, which was one or two years ago...

Quoted text here. Click to load it

The "best" filters are ones where a brute force, vary all components (say by  
a genetic algorithm for example) is used to satisfy the desired specs.

Most of those high brow "optimum" filter theorems are really just seeking an  
accomplice to filters that are not actually the real best. Its a "best in  
class" sort of thing, where the class is whatever it needs to be to have the  
claim "best".

I am aware of some work that had genetic algorithms re-discover the classis  
filters, and generate new, better ones that were, patentable.

-- Kevin Aylward
http://www.anasoft.co.uk - SuperSpice
http://www.kevinaylward.co.uk/ee/index.html


Re: fiddled filter design
On 11/09/2018 05:43 PM, Kevin Aylward wrote:

Quoted text here. Click to load it

A perfect delay, infinite bandwidth unity-gain filter sounds like a  
violation of conservation of energy to me. Maybe that's not what you  
meant tho

Quoted text here. Click to load it


Re: fiddled filter design


Quoted text here. Click to load it


Its called an ideal transmission line :-)

Somewhat annoyingly, recently discovered Cadence Virtuoso/Spectre have made  
what is, a standard Spice T line non functional. Replaced it with a bloody  
way too complicated mtline which for the life of me cant figure out how to  
set it up. As a stop gap, I used my SuperSpice automatic filter generator to  
generate a 4th order Bessel as a compromise. A/B ing with a TLine in SS  
shows really good phase matching.....


-- Kevin Aylward
http://www.anasoft.co.uk - SuperSpice
http://www.kevinaylward.co.uk/ee/index.html


Re: fiddled filter design
On 11/08/2018 05:42 PM, John Larkin wrote:
Quoted text here. Click to load it

I came up with a way of synthesizing a given input port parameter of  
generalized passive two-port impedance/admittance ladder networks a  
while back on paper which if i'm not wrong doesn't require futzing with  
that kinda ugly inelegant continued-fraction/polynomial expansion  
network synthesis method given in the textbooks, and seems more amenable  
to using divide-and-conquer/dynamic programming approach to synthesizing  
a response.

If there's demand for a proggie like that I will look into the method  
further didn't really have time to think it over more; I have no idea if  
the method is novel or not I couldn't find anything about it in what  
literature I looked at but there is a lot of EE literature and papers  
out there.

Quoted text here. Click to load it



Re: fiddled filter design
On Friday, November 9, 2018 at 7:54:21 AM UTC+11, John Larkin wrote:
Quoted text here. Click to load it

Williams and Taylor

https://www.amazon.com/Electronic-Filter-Handbook-McGraw-Hill-Handbooks/dp/0071471715

include "equiripple error" approximations to the classic filters, and discussion of making real filters for actual circuits.

"Diddling" parts values can be part of the design process - it doesn't oblige you to stop comprehending what you are doing.

--  
Bill Sloman, Sydney




Re: fiddled filter design
John Larkin wrote...
Quoted text here. Click to load it

 10kHz 5P filter.  I only see four poles.  How good is the 5kHz
 sine wave?  What's the filter's attenuation at say 15kHz?


--  
 Thanks,
    - Win

Re: fiddled filter design
wrote:

Quoted text here. Click to load it

The filter is down 52 dB at 45 KHz, which is the main DDS image. It's
actually up a half dB at 5K, our highest operating frequency. -3 at
7.5K.

Because we're running the DDS at 50 KHz, the spur at 45 KHz is already
down 20 dB from the 5 KHz fundamantal, which becomes -70 after
filtering. So the filter is overkill. The sine wave looks nice. DDS is
fun.

At 400 Hz, our main use case, the unfiltered DDS's big spur is already
(according to Spice) down 100 dB, not worth filtering. But I didn't
simulate the DAC quantization or the crummy DAC linearity, so it's not
that good in reality. The LPF won't help the DAC distortion at 400 Hz.

The downstream class-D power amp has some personality too, so we have
hooks in the uP code to add an amplitude tweak as a function of
frequency to flatten things if needed.  

This is a 30 VA sinewave source to excite synchros and resolvers and
LVDTs, so it really doesn't have to be very good. Synchros work off
400 Hz aircraft power, which is terrible, and "wild power" is worse.

This project got started two years ago. The amp kept blowing up and
the guy who designed it quit, so it fell to the bottom of the queue. I
have a new intern, so I figured it would be a good educational project
for him, and we'll finally get it done.


--  

John Larkin         Highland Technology, Inc

lunatic fringe electronics  


Re: fiddled filter design
John Larkin  
Quoted text here. Click to load it

Why so complicated?
Any 5.1 channel audio card would do, you can use 3 of the channels to make 3 phases if you need that.
No filtering needed, just a few lines of code to make the shifted sine waves.
If it is one phase only use sgen in Linux use sample rate 48kHz, clean sine waves.
Use a decent audio amp (analog) or audio amp IC.
There are many.
I drive my cryocooler at 60 Hz that way, 100 VA, as we only have 50 Hz mains.
Step up transformer is ring core main transformer in reverse,
For 400 Hz you can use a much smaller transformer.

Basic kids stuff.
  

Re: fiddled filter design
wrote:

Quoted text here. Click to load it

It will be a cute little box that complements our synchro/LVDT
products. It's not complicated. I'm not going to tell my customers to
get a PC and a sound card and a power amp and a transformer and
software so they don't have to buy my little box.

There are lots of 400 Hz power sources around, for aircraft
electronics testing, but most are in the KVA range and are big and
expensive, rackmount or worse. This will be a small benchtop unit.


--  

John Larkin         Highland Technology, Inc

lunatic fringe electronics  


Re: fiddled filter design
On a sunny day (Fri, 09 Nov 2018 09:19:34 -0800) it happened John Larkin

Quoted text here. Click to load it

If it is one phase, and < 80 VA, stick a raspberry in your box,
make it autostart sgen sending the 400 Hz to an amp like this:
  http://panteltje.com/panteltje/amplifier/index.html

I have this amp now for ? well you see how old the monitor is...
enough for your 30 VA.
Then you can have it play music too, connect to the internet shh
to it, remote service it, send software updates, etc etc.

Box can be very small, how much bigger depends on the step up transformer.
A raspberry pi (board) takes very little space, has stereo out, and uses hardly any power.
You do get about a minute of so boot-up time.
Oh and it has HDMI out and WiFi too these days.

 sgen  
sgen  Ver. 2.3.10 (May 2008)   Digital Signal Generator
Usage:  
 1: sgen [flags] waveform freq
      waveform is sine cosine square triangle sawtooth pulse noise off
 2: sgen [flags] sin|cos freq [phase]
      sin/cos has extra phase param (def. is 0 degrees)
 3: sgen [flags] pulse freq [Mark/Space]
      pulse has extra param Mark/Space % - def. is 10 (%)
Defaults: output continuously to /dev/dsp, 22050 samples/sec, mono,
          16 bit samples if possible, else 8 bit.
Default Config files are ".siggen.conf", "$(HOME)/.siggen.conf" and
"/etc/siggen.conf", searched in that order.
flags: -f,-a         force overwrite/append of/to file
       -o file       write digital sample to file ('-' is stdout)
       -w file       as '-o' but written as a wave file
       -C file       use file as local configuration file
       -s samples    generate with samplerate of samples/sec
       -v            be verbose.
 -8/-16 or -b 8|16   force 8 bit or 16 bit mode.
       -1,-2a        mono (def) or stereo in antiphase
       -A n          scale samples by n/100, def. n is 100
       -t N|Nm       play for N secs or Nm millisecs
       -x10|-x100    scale freqs down by factor of 10/100
                     allows freqs to 0.1/0.01 of a hertz.
~ # sgen -s 48000 sine 400


Re: fiddled filter design
wrote:

Quoted text here. Click to load it

That half-bridge config can have an interesting effect driving
inductive loads: it can take current from one supply and pump it into
the other. The victim rail can be pumped to a very high voltage.

We're using the TPA3251, which is a full-bridge class D amp. It's
under the heat sink, left side. It seems very well behaved.

https://www.dropbox.com/s/ejt7n10njhkotkb/P540_B_board.JPG?raw=1

Yes, that is a custom transformer. Sometimes you have to do that.

The little Talema ISDN transformer couples the voltage across a shunt
into the uP ADC, so we can compute and report the RMS output current.

Air enters the bottom of the box and gets exhaused from the top part
with a small fan. The notch in the board directs that air flow over
the amp.

We're seeing some 600 KHz ripple in the output, so we may hack in a
parallel LC trap or something.


--  

John Larkin         Highland Technology, Inc

lunatic fringe electronics  


Re: fiddled filter design
Quoted text here. Click to load it

Nice board, maybe better put the heatsink on the back?

Note the link I have to my old audio amp TDA7294,
is an analog amp, was discussed here long time ago,
it seems there is a follow up 'A'? version that is not so stable.
Anyways after years of abuse, driving big transformers, capacitive loads (piezo),
all sort of speakers, it is just my choice if I need some power to test something.
You say your amp broke down, I have the same experience with the TDA7492 class D 2 x 50 W amplifier from ebay:
 https://www.ebay.com/itm/141267388674
It died within a few days, do not even remember what I did with it,
tiny inductors it has.

And the ripple indeed, not something you want.

Of course class D is more efficient, but then if you need a new amp every few weeks it is not :-)



Quoted text here. Click to load it

Lemme see:
  http://panteltje.com/pub/TDA7294_amp_heatsink_IMG_6683.JPG


Quoted text here. Click to load it

Still, if you use a Raspberry Pi, and WiFi, somebody (not me, I cannot write Android apps) could
write an app to control it... seems these days everybody wants that, sales argument, and not class D with all that RF..
And then you could have it play 'star spanked? banner' or what was it? Or connect a monitor and play packman on it.
That sells.

Sorry, late here, need sleep.
:-)


Re: fiddled filter design
wrote:

Quoted text here. Click to load it


That's a problem with class-D switchers; not a problem with linear
amps. But linears can have huge power dissipations driving inductive
loads, which puts them into funny quadrants.


Quoted text here. Click to load it

The TI class D part has its power pad on the top side. That heat sink
arches over the chip and contacts it on top. We don't have room below
the board for the heat sink.



Quoted text here. Click to load it

The Gone Guy used  

MONOLITHIC POWER    MP7782DF-LF

which blew up any time it was in the mood, which was always.

Quoted text here. Click to load it

I got the TI eval board and abused it some. Never blew it up. It seems
solid.

Quoted text here. Click to load it

Our aerospace customers wouldn't like an RF link. They are paranoid
about any possibility of transporting any data out of a secure area.
If our products have any nonvolatile memory, there has to be a switch
to disable it so they can use it in secured areas.


--  

John Larkin         Highland Technology, Inc
picosecond timing   precision measurement  

We've slightly trimmed the long signature. Click to see the full one.
Re: fiddled filter design
Quoted text here. Click to load it


sigh, wiring is actually also ...
So PICs are out?
I assume you mean SDcards and other user writable stuff,
even to configure an FPGA you need FLASH.
Everybody already knows everything these days, they use ESP, advanced
neural nets that read what is playing in your brain.

Take that Boeing that went down last week or so,
from what I did read first thought was: HOW can a plane fall out of the sky?
Stall! Must be air speed sensor again....

You did not?, no you did not.
I mean save on spare parts, backup systems, I mean essential ones, gimme a break.

A 10$ GPS would have given them true speed to ground.
And so much for autonomous ...








Re: fiddled filter design
On 11/09/2018 12:19 PM, John Larkin wrote:
Quoted text here. Click to load it

Now you're thinking like a high-end audio designer! That's the spirit!


Re: fiddled filter design
On 11/09/2018 03:44 PM, bitrex wrote:
Quoted text here. Click to load it


Quoted text here. Click to load it

I worked in sales prior to working in engineering. "Sales prevention" is  
a frowned-upon activity!

Re: fiddled filter design

Quoted text here. Click to load it

Exactly. 30 watts, 250 Hz to 5 KHz, is enough for anybody.


--  

John Larkin         Highland Technology, Inc

lunatic fringe electronics  


Site Timeline