Prev: Alternative DMMs to Fluke?
Next: Can someone spare a spice model for a UC3842?
From: Rich Grise on 20 Apr 2006 18:31
On Thu, 20 Apr 2006 10:34:11 -0700, Jim Thompson wrote:

> On Thu, 20 Apr 2006 19:15:29 +0200, "maitre Aliboron"
> <aliboron(a)despammed.com> wrote:
>
>>> Use an external accurate resistor ?:-)
>>
>>impossible... :-))
>>
>>For the moment I'm oriented toward something weird and
>>not usual like CCII, differential transconductors or very nice
>>stuff like that... But I guess I can find something simpler, (I
>>highly hope).
>
> How can you make an accurate absolute value?
>
> I _have_ done systems that periodically re-calibrate themselves, but
> you have to feed it a reference during the re-calibration.
>
> ...Jim Thompson

Speaking of self-calibration, I wonder what ever happened to that
self-tuning piano?

Thanks!
Rich


From: Don Lancaster on 20 Apr 2006 19:03
Rich Grise wrote:
> On Thu, 20 Apr 2006 10:34:11 -0700, Jim Thompson wrote:
>
>
>>On Thu, 20 Apr 2006 19:15:29 +0200, "maitre Aliboron"
>><aliboron(a)despammed.com> wrote:
>>
>>
>>>>Use an external accurate resistor ?:-)
>>>
>>>impossible... :-))
>>>
>>>For the moment I'm oriented toward something weird and
>>>not usual like CCII, differential transconductors or very nice
>>>stuff like that... But I guess I can find something simpler, (I
>>>highly hope).
>>
>>How can you make an accurate absolute value?
>>
>>I _have_ done systems that periodically re-calibrate themselves, but
>>you have to feed it a reference during the re-calibration.
>>
>> ...Jim Thompson
>
>
> Speaking of self-calibration, I wonder what ever happened to that
> self-tuning piano?
>
> Thanks!
> Rich
>
>

Pianos are enormously difficult to tune because the overtones are NON
harmonic! The keyboard needs "stretched" to compensate for this effect.

Apparently the lateral stiffness of the strings moves the seventh
ovretone up to where the eighth harmonic is supposed to be.

The lowest notes are lowered by as much as thirty cents.
The highest notes are raised by as much as twelve cents.

A piano tuned with a frequency meter will sound awful.

And Opporknockity tunes but once.

--
Many thanks,

Don Lancaster voice phone: (928)428-4073
Synergetics 3860 West First Street Box 809 Thatcher, AZ 85552
rss: http://www.tinaja.com/whtnu.xml email: don(a)tinaja.com

Please visit my GURU's LAIR web site at http://www.tinaja.com
From: Winfield Hill on 20 Apr 2006 19:27
maitre Aliboron wrote...
>
>> Use an external accurate resistor ?:-)
>
> impossible... :-))
>
> For the moment I'm oriented toward something weird
> and not usual like CCII, differential transconductors
> or very nice stuff like that... But I guess I can
> find something simpler, (I highly hope).

I don't see how that would help, unless you have a
rather special situation we don't see.

As for an accurate voltage-controlled current source,
your MOS IC can be calibrated during test, with any
of several special features you add to the design:
laser-trimmed or zener-zapped resistor segments,
EEPROM DAC-style calibration, EEPROM uP self-cal, etc.

These days, we are told, such enhancements take very
little silicon real estate, and may even be plunked-in
from IP libraries. The EEPROM approach may be more
attractive because it's practical for more fab lines.

Alternately you could design your IC so the current-
calibration feature works in conjunction with a uP,
which would determine and store the cal coefficients
itself under software control, and feed them to your
chip at powerup. :-) In this way you can avoid a
large painful expensive external precision resistor.

Jim can fill you in.


--
Thanks,
- Win
From: Michael A. Terrell on 20 Apr 2006 21:56
Don Lancaster wrote:
>
> And Opporknockity tunes but once.


One of Archie Campbell's spoonerisms and a story he told a number of
times on the "Grand Ole Opry", but he said, "Opporknockity only tunes
once!". ;-)


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida
From: maitre Aliboron on 21 Apr 2006 03:17
> Does it all have to be on-chip?

yes, that's the problem

> Can you make accurate capacitors?

yes/no. Process is an unexpensive standard CMOS
with no options (another big constraint). The only capa
available is a MOS capacitor, not really the best
capacitor in the world, but we can survive with...

Are you thinking to something like "switched capacitors
(active) current sources"?

Not bad, though I need at least an oscillator.
I was looking for simpler solutions, (I know, It's
not easy...)

Thanks anyway.

maitre Aliboron




First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4 5
Prev: Alternative DMMs to Fluke?
Next: Can someone spare a spice model for a UC3842?