On May 1, 10:57&#4294967295;pm, "steveu" <steveu@n_o_s_p_a_m.coppice.org> wrote:

  ...

> >Elderly "sanity check". IIRC :)
> > &#4294967295; Did not mechanical multi blade capacitors have 'notches' in
> > &#4294967295; some blades?
>
> Many had very strange shapes, often with a few small holes in the blades.
> They seemed more like they were designed for artistic effect than for any
> engineering goal.

That strangeness was at least in some cases a result of empiricism.
When I worked for Andrea Radio in the '50s, we built a TV alignment
generator under contract for RCA (their logo). Here, the frequency
markings and actual frequencies had to match quite well. Originally,
the silk screening was drawn to match reality. On one batch, the
calibration was way off. RCA agreed to pay extra for hand trimming
rather than take delivery and fix it themselves. (Our contract called
for assembly, not design.) I was able to reshape one of the plates to
achieve alignment without hand trimming, but talk about strange shape!

Jerry
--
Engineering is the art of making what you want from things you can get.

>Jerry Avins wrote:
>> On May 1, 4:56 am, Tauno Voipio<tauno.voi...@notused.fi.invalid>
>> wrote:
>>> On 1.5.11 11:39 , John Monro wrote:
>>>
>>>> I noticed that when the first digitally controlled attenuator chips
came
>>>> out they had linear steps. I assumed at the time that it was simpler
to
>>>> lay out the chip that way.
>>>
>>> The first Japanese transistor sets had quasi-log audio potentiometers:
>>> there was two linear sections with an annoying knee joining them.
>>>
>>>> Another mystery (to me) is why radio manufacturers stopped using
>>>> square-law tuning capacitors on AM radios. This happened around the
time
>>>> that transistor radios came on the scene.
>>>
>>> My guess is that the question is about space inside the box.
>>> The square-law capacitors need much more free space around
>>> the open capacitor than a cap with semi-circular rotor.
>>
>> Yes.
>>
>>>> The old valve AM radios had nice, linear frequency calibrations
whereas
>>>> the consumer transistor AM radios around now have nasty little tuning
>>>> capacitors with semi-circular vanes, causing the stations to be
squeezed
>>>> up at the high-frequency end of the dial on the AM band and spread
out
>>>> along the low-frequency end.
>>>
>>> An old venerable tube AM set was a PITA to linearize by twisting
>>> the outer vanes to make the stations sit correctly on the scale.
>>> If the scale is squeezed there is no reason to do the time-consuming
>>> linearity tuning of the oscillator section.
>>
>> No. The outer vanes were adjusted to track the tuning of the RF
>> amplifier and the LO. This was cheaper and more exact than the padder-
>> cum-trimmer approach and not much more difficult to trim.
>>
>> Jerry
>>
>
>Elderly "sanity check". IIRC :)
>   Did not mechanical multi blade capacitors have 'notches' in
>   some blades?

Many had very strange shapes, often with a few small holes in the blades.
They seemed more like they were designed for artistic effect than for any
engineering goal.

Steve


>A half-century ago, I wasn't interested in any linearity between 
>"dial markings" and "dial rotation".

On May 1, 4:41&#4294967295;pm, Tauno Voipio <tauno.voi...@notused.fi.invalid>
wrote:
> On 1.5.11 6:29 , Jerry Avins wrote:
>
>
>
>
>
>
>
>
>
> > On May 1, 4:56 am, Tauno Voipio<tauno.voi...@notused.fi.invalid>
> > wrote:
> >> On 1.5.11 11:39 , John Monro wrote:
>
> >>> I noticed that when the first digitally controlled attenuator chips came
> >>> out they had linear steps. I assumed at the time that it was simpler to
> >>> lay out the chip that way.
>
> >> The first Japanese transistor sets had quasi-log audio potentiometers:
> >> there was two linear sections with an annoying knee joining them.
>
> >>> Another mystery (to me) is why radio manufacturers stopped using
> >>> square-law tuning capacitors on AM radios. This happened around the time
> >>> that transistor radios came on the scene.
>
> >> My guess is that the question is about space inside the box.
> >> The square-law capacitors need much more free space around
> >> the open capacitor than a cap with semi-circular rotor.
>
> > Yes.
>
> >>> The old valve AM radios had nice, linear frequency calibrations whereas
> >>> the consumer transistor AM radios around now have nasty little tuning
> >>> capacitors with semi-circular vanes, causing the stations to be squeezed
> >>> up at the high-frequency end of the dial on the AM band and spread out
> >>> along the low-frequency end.
>
> >> An old venerable tube AM set was a PITA to linearize by twisting
> >> the outer vanes to make the stations sit correctly on the scale.
> >> If the scale is squeezed there is no reason to do the time-consuming
> >> linearity tuning of the oscillator section.
>
> > No. The outer vanes were adjusted to track the tuning of the RF
> > amplifier and the LO. This was cheaper and more exact than the padder-
> > cum-trimmer approach and not much more difficult to trim.
>
> > Jerry
> > --
> > Engineering is the art of making what you want from things you can
> > get.
>
> It was for both. The oscillator side outer vanes were
> twisted to set the linearity, and the antenna section
> vanes (and maybe RF amplifier section) vanes were twisted
> to make the tracking. The fromt end tuning was quite
> broad in parctice, so the linearization was more important.
>
> It was still usual to have some of the padding capacitor
> in series with the oscillator section.

Not in the cheaper "all-American five".

> Been there, done that

"Those were the days, my friend!"

Jerry
--
Engineering is the art of making what you want from things you can get.

On May 1, 3:48&#4294967295;pm, Richard Owlett <rowl...@pcnetinc.com> wrote:
> Jerry Avins wrote:
> > On May 1, 4:56 am, Tauno Voipio<tauno.voi...@notused.fi.invalid>
> > wrote:
> >> On 1.5.11 11:39 , John Monro wrote:
>
> >>> I noticed that when the first digitally controlled attenuator chips came
> >>> out they had linear steps. I assumed at the time that it was simpler to
> >>> lay out the chip that way.
>
> >> The first Japanese transistor sets had quasi-log audio potentiometers:
> >> there was two linear sections with an annoying knee joining them.
>
> >>> Another mystery (to me) is why radio manufacturers stopped using
> >>> square-law tuning capacitors on AM radios. This happened around the time
> >>> that transistor radios came on the scene.
>
> >> My guess is that the question is about space inside the box.
> >> The square-law capacitors need much more free space around
> >> the open capacitor than a cap with semi-circular rotor.
>
> > Yes.
>
> >>> The old valve AM radios had nice, linear frequency calibrations whereas
> >>> the consumer transistor AM radios around now have nasty little tuning
> >>> capacitors with semi-circular vanes, causing the stations to be squeezed
> >>> up at the high-frequency end of the dial on the AM band and spread out
> >>> along the low-frequency end.
>
> >> An old venerable tube AM set was a PITA to linearize by twisting
> >> the outer vanes to make the stations sit correctly on the scale.
> >> If the scale is squeezed there is no reason to do the time-consuming
> >> linearity tuning of the oscillator section.
>
> > No. The outer vanes were adjusted to track the tuning of the RF
> > amplifier and the LO. This was cheaper and more exact than the padder-
> > cum-trimmer approach and not much more difficult to trim.
>
> > Jerry
>
> Elderly "sanity check". IIRC :)
> &#4294967295; &#4294967295;Did not mechanical multi blade capacitors have 'notches' in
> &#4294967295; &#4294967295;some blades?


Yes. Bending the individual pieces was how the capacitance was varied
from the production standard.

> A half-century ago, I wasn't interested in any linearity between
> "dial markings" and "dial rotation".

Except for communications receivers, approximate linearity was all
anyone asked for.

Jerry

On 1.5.11 6:29 , Jerry Avins wrote:
> On May 1, 4:56 am, Tauno Voipio<tauno.voi...@notused.fi.invalid>
> wrote:
>> On 1.5.11 11:39 , John Monro wrote:
>>
>>> I noticed that when the first digitally controlled attenuator chips came
>>> out they had linear steps. I assumed at the time that it was simpler to
>>> lay out the chip that way.
>>
>> The first Japanese transistor sets had quasi-log audio potentiometers:
>> there was two linear sections with an annoying knee joining them.
>>
>>> Another mystery (to me) is why radio manufacturers stopped using
>>> square-law tuning capacitors on AM radios. This happened around the time
>>> that transistor radios came on the scene.
>>
>> My guess is that the question is about space inside the box.
>> The square-law capacitors need much more free space around
>> the open capacitor than a cap with semi-circular rotor.
>
> Yes.
>
>>> The old valve AM radios had nice, linear frequency calibrations whereas
>>> the consumer transistor AM radios around now have nasty little tuning
>>> capacitors with semi-circular vanes, causing the stations to be squeezed
>>> up at the high-frequency end of the dial on the AM band and spread out
>>> along the low-frequency end.
>>
>> An old venerable tube AM set was a PITA to linearize by twisting
>> the outer vanes to make the stations sit correctly on the scale.
>> If the scale is squeezed there is no reason to do the time-consuming
>> linearity tuning of the oscillator section.
>
> No. The outer vanes were adjusted to track the tuning of the RF
> amplifier and the LO. This was cheaper and more exact than the padder-
> cum-trimmer approach and not much more difficult to trim.
>
> Jerry
> --
> Engineering is the art of making what you want from things you can
> get.


It was for both. The oscillator side outer vanes were
twisted to set the linearity, and the antenna section
vanes (and maybe RF amplifier section) vanes were twisted
to make the tracking. The fromt end tuning was quite
broad in parctice, so the linearization was more important.

It was still usual to have some of the padding capacitor
in series with the oscillator section.

-- 

Been there, done that
Tauno Voipio

Jerry Avins wrote:
> On May 1, 4:56 am, Tauno Voipio<tauno.voi...@notused.fi.invalid>
> wrote:
>> On 1.5.11 11:39 , John Monro wrote:
>>
>>> I noticed that when the first digitally controlled attenuator chips came
>>> out they had linear steps. I assumed at the time that it was simpler to
>>> lay out the chip that way.
>>
>> The first Japanese transistor sets had quasi-log audio potentiometers:
>> there was two linear sections with an annoying knee joining them.
>>
>>> Another mystery (to me) is why radio manufacturers stopped using
>>> square-law tuning capacitors on AM radios. This happened around the time
>>> that transistor radios came on the scene.
>>
>> My guess is that the question is about space inside the box.
>> The square-law capacitors need much more free space around
>> the open capacitor than a cap with semi-circular rotor.
>
> Yes.
>
>>> The old valve AM radios had nice, linear frequency calibrations whereas
>>> the consumer transistor AM radios around now have nasty little tuning
>>> capacitors with semi-circular vanes, causing the stations to be squeezed
>>> up at the high-frequency end of the dial on the AM band and spread out
>>> along the low-frequency end.
>>
>> An old venerable tube AM set was a PITA to linearize by twisting
>> the outer vanes to make the stations sit correctly on the scale.
>> If the scale is squeezed there is no reason to do the time-consuming
>> linearity tuning of the oscillator section.
>
> No. The outer vanes were adjusted to track the tuning of the RF
> amplifier and the LO. This was cheaper and more exact than the padder-
> cum-trimmer approach and not much more difficult to trim.
>
> Jerry
>

Elderly "sanity check". IIRC :)
   Did not mechanical multi blade capacitors have 'notches' in
   some blades?

A half-century ago, I wasn't interested in any linearity between 
"dial markings" and "dial rotation".

On May 1, 4:56&#4294967295;am, Tauno Voipio <tauno.voi...@notused.fi.invalid>
wrote:
> On 1.5.11 11:39 , John Monro wrote:
>
> > I noticed that when the first digitally controlled attenuator chips came
> > out they had linear steps. I assumed at the time that it was simpler to
> > lay out the chip that way.
>
> The first Japanese transistor sets had quasi-log audio potentiometers:
> there was two linear sections with an annoying knee joining them.
>
> > Another mystery (to me) is why radio manufacturers stopped using
> > square-law tuning capacitors on AM radios. This happened around the time
> > that transistor radios came on the scene.
>
> My guess is that the question is about space inside the box.
> The square-law capacitors need much more free space around
> the open capacitor than a cap with semi-circular rotor.

Yes.

> > The old valve AM radios had nice, linear frequency calibrations whereas
> > the consumer transistor AM radios around now have nasty little tuning
> > capacitors with semi-circular vanes, causing the stations to be squeezed
> > up at the high-frequency end of the dial on the AM band and spread out
> > along the low-frequency end.
>
> An old venerable tube AM set was a PITA to linearize by twisting
> the outer vanes to make the stations sit correctly on the scale.
> If the scale is squeezed there is no reason to do the time-consuming
> linearity tuning of the oscillator section.

No. The outer vanes were adjusted to track the tuning of the RF
amplifier and the LO. This was cheaper and more exact than the padder-
cum-trimmer approach and not much more difficult to trim.

Jerry
--
Engineering is the art of making what you want from things you can
get.

On 1.5.11 11:39 , John Monro wrote:
> I noticed that when the first digitally controlled attenuator chips came
> out they had linear steps. I assumed at the time that it was simpler to
> lay out the chip that way.

The first Japanese transistor sets had quasi-log audio potentiometers:
there was two linear sections with an annoying knee joining them.

> Another mystery (to me) is why radio manufacturers stopped using
> square-law tuning capacitors on AM radios. This happened around the time
> that transistor radios came on the scene.

My guess is that the question is about space inside the box.
The square-law capacitors need much more free space around
the open capacitor than a cap with semi-circular rotor.

> The old valve AM radios had nice, linear frequency calibrations whereas
> the consumer transistor AM radios around now have nasty little tuning
> capacitors with semi-circular vanes, causing the stations to be squeezed
> up at the high-frequency end of the dial on the AM band and spread out
> along the low-frequency end.

An old venerable tube AM set was a PITA to linearize by twisting
the outer vanes to make the stations sit correctly on the scale.
If the scale is squeezed there is no reason to do the time-consuming
linearity tuning of the oscillator section.

-- 

Tauno Voipio, OH2UG (since 50 years)

I noticed that when the first digitally controlled 
attenuator chips came out they had linear steps.  I assumed 
at the time that it was simpler to lay out the chip that way.

Another mystery (to me) is why radio manufacturers stopped 
using square-law tuning capacitors on AM radios.  This 
happened around the time that transistor radios came on the 
scene.

The old valve AM radios had nice, linear frequency 
calibrations whereas the consumer transistor AM radios 
around now have nasty little tuning capacitors with 
semi-circular vanes, causing the stations to be squeezed up 
at the high-frequency end of the dial on the AM band and 
spread out along the low-frequency end.

Regards,
John

On 1/05/2011 1:36 PM, Jerry Avins wrote:
> Why do most software audio volume controls have linear tapers? I would
> have guesses that simulated log-taper controls would be standard.
>
> Jerry
> --
> Engineering is the art of making what you want from things you can get.