Analogue moving coil meter range extension?

[T i m]

Hi All,

Not a 'repair' question as such but I was thinking of using an old
skool 250uA FSD moving coil panel meter to read the 10 to 15V DC range
(a lead acid battery charge / discharge indicator), next to a 0-30A
Ammeter (inc shunt etc).

I was thinking of using either a voltage reference such as
LM4040DIZ-10.0 or could I get away with a simple zener (BZX55C?) with
associated resistor(s) please?

Once I have the 10V (that might need it's own drive resistor?) for the
-ve side of the meter I think I'd need a 20k resistor in series with
the meter to my 10-15V supply (a lead acid battery)?

R = V/I = 5 / 250 x 10^-6 = 20k ohms.

Or ... I could use a straight potential divider on the grounds that a
moving coil meter isn't going to be that accurate in the first place?

Maybe: -Ve, 10k, 5k, +Ve with 20k in series with the meter across the
5K?

Am I close please? ;-)

Cheers, T i m

[pf...@aol.com]

There should be a way to rig a simple resistor to do what you need. You may have to re-scale the meter, but that should do it. Perhaps an adjustable wire-wound resistor that you could tweak until *your* voltage range matched the meter range?

https://www.alliedelec.com/search/productview.aspx?SKU=70023692 As a one-off, the cost would not be too excessive. And the meter would be just as 'accurate' as before. But there should be no reason to get fancy. It would all depend on how much current is going through the resistor(s).

Peter Wieck
Melrose Park, PA

[tabb...@gmail.com]

How accurate do you want the readings? Add your errors together. MC meters are usually pretty good on accuracy.


NT

[T i m]

On Thu, 21 Sep 2017 11:30:17 -0700 (PDT), "pf...@aol.com"
<pf...@aol.com> wrote:

>There should be a way to rig a simple resistor to do what you need.

I put forward 3 ways that were pretty simple. ;-)

> You may have to re-scale the meter, but that should do it.

I might adjust the scale reading from 1 to 5 to 10 to 15 but I didn't
really want to do much more.

> Perhaps an adjustable wire-wound resistor that you could tweak until *your* voltage range matched the meter range?

The problem with any variable resistor is stability. I could be of for
very fine tuning though.
>
>https://www.alliedelec.com/search/productview.aspx?SKU=70023692

That's a bigun. ;-)

>As a one-off, the cost would not be too excessive.

Ok. Ironically the moving coil meter itself is more expensive than a
digital display or even a wattmeter but I actually want to go old
skool here.

> And the meter would be just as 'accurate' as before.

Ok.

> But there should be no reason to get fancy.

No, no reason maybe but if going a 'bit' fancy gives me a better
solution and not for a lot of money or effort I don't mind going for
it. ;-)

>It would all depend on how much current is going through the resistor(s).

Hopefully, no more than the 250uA required for FSD Peter. ;-)

Cheers, T i m

[T i m]

On Thu, 21 Sep 2017 12:05:59 -0700 (PDT), tabb...@gmail.com wrote:

<snip>

>
>How accurate do you want the readings?

As accurate as the meter can be itself, ideally.

>Add your errors together. MC meters are usually pretty good on accuracy.

Ok, hopefully the biggest issue will be parallax error and my old
eyes. ;-)

Cheers, T i m

[tabb...@gmail.com]

On Thursday, 21 September 2017 20:24:48 UTC+1, T i m wrote:

> On Thu, 21 Sep 2017 12:05:59 -0700 (PDT), tabbypurr wrote:
>
> <snip>
>
> >
> >How accurate do you want the readings?
>
> As accurate as the meter can be itself, ideally.

Your total error is all the errors added together, so you'd need some exceptional circuitry to not add anything significant to the meter's error, which might be 1% or so. More realistic would be to keep the electronic errors within 1%.

To get that you'd be using an opamp to knock 10v off, with 0.1% resistors everywhere it matters. Avoiding anything getting hot always helps keep the errors down, so anything that gets warm heatsink it, and operate parts way under their limits. Not that that will be a problem with this meter.


NT

[tabb...@gmail.com]

On Thursday, 21 September 2017 20:22:43 UTC+1, T i m wrote:
> On Thu, 21 Sep 2017 11:30:17 -0700 (PDT), "pf...@aol.com"
> <pf...@aol.com> wrote:
>
> >There should be a way to rig a simple resistor to do what you need.

there isn't

> I put forward 3 ways that were pretty simple. ;-)
>
> > You may have to re-scale the meter, but that should do it.
>
> I might adjust the scale reading from 1 to 5 to 10 to 15 but I didn't
> really want to do much more.
>
> > Perhaps an adjustable wire-wound resistor that you could tweak until *your* voltage range matched the meter range?

won't get you 10-15v range

> The problem with any variable resistor is stability. I could be of for
> very fine tuning though.
> >
> >https://www.alliedelec.com/search/productview.aspx?SKU=70023692
>
> That's a bigun. ;-)
>
> >As a one-off, the cost would not be too excessive.
>
> Ok. Ironically the moving coil meter itself is more expensive than a
> digital display or even a wattmeter but I actually want to go old
> skool here.
>
> > And the meter would be just as 'accurate' as before.

impossible

> Ok.
>
> > But there should be no reason to get fancy.
>
> No, no reason maybe but if going a 'bit' fancy gives me a better
> solution and not for a lot of money or effort I don't mind going for
> it. ;-)
>
> >It would all depend on how much current is going through the resistor(s).
>
> Hopefully, no more than the 250uA required for FSD Peter. ;-)
>
> Cheers, T i m

:) Your opamp should do a nice job of protecting the meter, you just need to protect the opamp.


NT

[pf...@aol.com]

Like this?

http://thumbs.ebaystatic.com/images/g/XAMAAOSw-0xYRaJT/s-l225.jpg

Right off the shelf.

Several others from similar sources with a table of resistance values to make the desired range (in volts) and either blank faces or sequential numeric faces. Sorry for pointing to an Euro-meter, but just that they are out there.

Had I known, the "meter man" at Kutztown would have had exactly what you want in the size you want it, new or used. He has hundreds. Now you have to wait until spring.

[Ian Field]

"T i m" <ne...@spaced.me.uk> wrote in message
news:vas7scdd02ptrj9ig...@4ax.com...

> Hi All,
>
> Not a 'repair' question as such but I was thinking of using an old
> skool 250uA FSD moving coil panel meter to read the 10 to 15V DC range
> (a lead acid battery charge / discharge indicator), next to a 0-30A
> Ammeter (inc shunt etc).
>
> I was thinking of using either a voltage reference such as
> LM4040DIZ-10.0 or could I get away with a simple zener (BZX55C?) with
> associated resistor(s) please?

Most people just use a Zener and maybe a few normal (or SB) forward volt
drops to trim the offset.

I was thinking the TL431 programmable Zener could be calibrated to exactly
what you need - but it needs a minimum current through it (possibly 1mA or
so).

You can probably incorporate a TL431 into some kind of bridge circuit with
the movement spanning the 2 arms, but it will almost certainly need some
kind of power supply.

[T i m]

On Thu, 21 Sep 2017 12:35:16 -0700 (PDT), tabb...@gmail.com wrote:

>On Thursday, 21 September 2017 20:24:48 UTC+1, T i m wrote:
>> On Thu, 21 Sep 2017 12:05:59 -0700 (PDT), tabbypurr wrote:
>>
>> <snip>
>>
>> >
>> >How accurate do you want the readings?
>>
>> As accurate as the meter can be itself, ideally.
>
>Your total error is all the errors added together,

I know.

>so you'd need some exceptional circuitry to not add anything significant to the meter's error, which might be 1% or so. More realistic would be to keep the electronic errors within 1%.

Understood.

>
>To get that you'd be using an opamp to knock 10v off,

Why not a zener or precision voltage IC?

>with 0.1% resistors everywhere it matters.

As long as the final result calibrates correctly, does it matter what
the tolerances are / were?

>Avoiding anything getting hot always helps keep the errors down,

With 250uA FSD I can't anything much getting warm?

>so anything that gets warm heatsink it, and operate parts way under their limits. Not that that will be a problem with this meter.

Quite.

So, precision reference, zener, resistor bridge?

With a precision reference and zener my meter wouldn't start doing
anything till the measured voltage started to exceed 10V, even if I
was ramping from 0 to 15V (and that's fine for a 12V lead acid).

With a resistor bridge with a 1/3 split I should be able to get FSD at
my 15V but anything less wouldn't work (would it?) ... so it's going
to have to be either of the first two ... or something else?

Basically, as long as I hold the negative of the meter at the same
voltage as the positive at anything 10V and below, the meter won't do
anything. Then as the +ve of the meter exceeds 10V and goes to 15V, a
meter set to read 0-5 volts should follow ok, giving me reasonable /
best possible accuracy across the working range of the battery.

The only prerequisites are to ensure the -ve of the meter is held
exactly at +10V (from a supply ranging from 10.5 to 14.4V) and that
the meter current limiting resistor is correctly matched (to give the
250uA required for FSD at 5V PD).

As long as any reference would be suitably driving at anything above
10V and not overdriven at 15V then we should be good to go?

Cheers, T i m

[T i m]

On Thu, 21 Sep 2017 21:43:31 +0100, "Ian Field"
<gangprobi...@virginmedia.com> wrote:

>
>
>"T i m" <ne...@spaced.me.uk> wrote in message
>news:vas7scdd02ptrj9ig...@4ax.com...
>> Hi All,
>>
>> Not a 'repair' question as such but I was thinking of using an old
>> skool 250uA FSD moving coil panel meter to read the 10 to 15V DC range
>> (a lead acid battery charge / discharge indicator), next to a 0-30A
>> Ammeter (inc shunt etc).
>>
>> I was thinking of using either a voltage reference such as
>> LM4040DIZ-10.0 or could I get away with a simple zener (BZX55C?) with
>> associated resistor(s) please?
>
>Most people just use a Zener and maybe a few normal (or SB) forward volt
>drops to trim the offset.

Ok.

>
>I was thinking the TL431 programmable Zener could be calibrated to exactly
>what you need - but it needs a minimum current through it (possibly 1mA or
>so).

Well that's ok?

>
>You can probably incorporate a TL431 into some kind of bridge circuit with
>the movement spanning the 2 arms, but it will almost certainly need some
>kind of power supply.

Because the reference voltage will be too close to the minimum voltage
expected Ian?

If that is the case I could use a very light bang converter to bring
the supply up to say 15 V and power it from the source that is being
monitored in use (running an electric outboard from a 93Ah battery).

Cheers, T i m

[tabb...@gmail.com]

On Thursday, 21 September 2017 21:44:35 UTC+1, T i m wrote:

> On Thu, 21 Sep 2017 12:35:16 -0700 (PDT), tabbypurr wrote:
> >On Thursday, 21 September 2017 20:24:48 UTC+1, T i m wrote:
> >> On Thu, 21 Sep 2017 12:05:59 -0700 (PDT), tabbypurr wrote:
> >>
> >> <snip>
> >>
> >> >
> >> >How accurate do you want the readings?
> >>
> >> As accurate as the meter can be itself, ideally.
> >
> >Your total error is all the errors added together,
>
> I know.
>
> >so you'd need some exceptional circuitry to not add anything significant to the meter's error, which might be 1% or so. More realistic would be to keep the electronic errors within 1%.
>
> Understood.
> >
> >To get that you'd be using an opamp to knock 10v off,
>
> Why not a zener or precision voltage IC?

which one?

> >with 0.1% resistors everywhere it matters.
>
> As long as the final result calibrates correctly, does it matter what
> the tolerances are / were?

There are 2 ways to get accuracy.
1. adjust/calibrate
2. build it accurately
I don't know which suits you better.
Long ago I used to use selected 5% resistors where 1% were wanted, but they don't stay as accurate as 1%ers.

> >Avoiding anything getting hot always helps keep the errors down,
>
> With 250uA FSD I can't anything much getting warm?

no :)

> >so anything that gets warm heatsink it, and operate parts way under their limits. Not that that will be a problem with this meter.
>
> Quite.
>
> So, precision reference, zener, resistor bridge?
>
> With a precision reference and zener my meter wouldn't start doing
> anything till the measured voltage started to exceed 10V, even if I
> was ramping from 0 to 15V (and that's fine for a 12V lead acid).

If you had a perfect-knee perfectly accurate voltage zener then sure, but where will you find one of those?

> With a resistor bridge with a 1/3 split I should be able to get FSD at
> my 15V but anything less wouldn't work (would it?) ... so it's going
> to have to be either of the first two ... or something else?

or resistors on the input of an opamp so it only turns on once 10v is reached

> Basically, as long as I hold the negative of the meter at the same
> voltage as the positive at anything 10V and below, the meter won't do
> anything. Then as the +ve of the meter exceeds 10V and goes to 15V, a
> meter set to read 0-5 volts should follow ok, giving me reasonable /
> best possible accuracy across the working range of the battery.
>
> The only prerequisites are to ensure the -ve of the meter is held
> exactly at +10V (from a supply ranging from 10.5 to 14.4V) and that
> the meter current limiting resistor is correctly matched (to give the
> 250uA required for FSD at 5V PD).
>
> As long as any reference would be suitably driving at anything above
> 10V and not overdriven at 15V then we should be good to go?
>
> Cheers, T i m

There's always protection to consider.


NT

[Foxs Mercantile]

Sigh....

The simple answer is put a 62K 5% resistor in series with the meter.
It's read about 4% low (240 uA instead of 250 uA.)

You wanna get a little closer? put a 100K and a 150K in parallel.
Wanna make it adjustable? Put a 100K trimpot in parallel with a 180K
resistor.

This ain't rocket science folks.

--
Jeff-1.0
wa6fwi
http://www.foxsmercantile.com

[T i m]

On Thu, 21 Sep 2017 16:54:07 -0500, Foxs Mercantile <jda...@att.net>
wrote:

>Sigh....
>
>The simple answer is put a 62K 5% resistor in series with the meter.
>It's read about 4% low (240 uA instead of 250 uA.)

Ok, so with our source at 15V we get 240 uA and ~FSD on our meter.
That's 14.88V across the resistor and 0.12V across the meter,
therefore the meter is ~500 ohms?

At 10V we get 2/3 of 240 uA (160uA) and so our voltage across the
meter is now 0.0256 (~1/5th of FSD) and so the meter will read 1 volt
and not 0?

>
>You wanna get a little closer? put a 100K and a 150K in parallel.
>Wanna make it adjustable? Put a 100K trimpot in parallel with a 180K
>resistor.

The problem is that the voltage across the meter will fall in
proportion to the voltage across the entire network, not just across
the meter component (which is why I ruled it out in my second post).
;-)

>
>This ain't rocket science folks.

Maybe not but can appear as such to some (I think). ;-)

Cheers, T i m

[Foxs Mercantile]

On 9/21/2017 5:41 PM, T i m wrote:
> On Thu, 21 Sep 2017 16:54:07 -0500, Foxs Mercantile <jda...@att.net>
> wrote:
>
>> Sigh....
>>
>> The simple answer is put a 62K 5% resistor in series with the meter.
>> It's read about 4% low (240 uA instead of 250 uA.)
>
> Ok, so with our source at 15V we get 240 uA and ~FSD on our meter.
> That's 14.88V across the resistor and 0.12V across the meter,
> therefore the meter is ~500 ohms?

No that's 15 volt across the resistor only so it's ~240 uA.
With the internal meter resistance it would be slightly less current.

> At 10V we get 2/3 of 240 uA (160uA) and so our voltage across the
> meter is now 0.0256 (~1/5th of FSD) and so the meter will read 1 volt
> and not 0?

No, it's still assuming the meter resistance is close to zero.
at 160 uA, it will read 2/3 scale at 10 volts.

>> You wanna get a little closer? put a 100K and a 150K in parallel.
>> Wanna make it adjustable? Put a 100K trimpot in parallel with a 180K
>> resistor.
>
> The problem is that the voltage across the meter will fall in
> proportion to the voltage across the entire network, not just across
> the meter component (which is why I ruled it out in my second post).

No, you're NOT measuring voltage across the meter. You're measuring
current through the meter.

It's a series circuit. (The meter and the series resistor.) As such
in this case, the series resistance is much higher than the internal
resistance of the meter, and that defines what the current will be for
a given voltage.

I.e. 60K ohm, 15 volts, 250 micro-amps.

[Foxs Mercantile]

How things work:
Amp meters, by definition are suppose to be low resistance items.
When you're measuring current through a circuit, you don't want
to waste any of the available voltage across the meter.

As an example of this, current shunts for meters are usually rated
at X-amps for 50-100 mV across the shunt.

On the other hand, volt meters are put in parallel with a load, and
to be accurate, they should be many time higher than the load
resistance so they don't pull the supply voltage down that you're
trying to measure.

Back in the "good old days" for example, they would specify taking
a voltage reading with a "minimum 20K ohms/volt" meter.
If your trying to measure 100 volts, if you have a 150 volt range
on the meter, this would be a load resistance of 3 megohms across
the circuit you're measuring.

In the example given for a 250 uA meter, that's only 4K/volt.
But for measuring 10-15 volts, most supplies are good for anywhere
from 1-25 amps, so the extra 250 uA is hardly noticeable.

[tabb...@gmail.com]

On Thursday, 21 September 2017 22:54:17 UTC+1, Foxs Mercantile wrote:
> Sigh....
>
> The simple answer is put a 62K 5% resistor in series with the meter.
> It's read about 4% low (240 uA instead of 250 uA.)
>
> You wanna get a little closer? put a 100K and a 150K in parallel.
> Wanna make it adjustable? Put a 100K trimpot in parallel with a 180K
> resistor.
>
> This ain't rocket science folks.

It's also plainly noncompliant with the project requirements.


NT

[T i m]

On Thu, 21 Sep 2017 14:37:35 -0700 (PDT), tabb...@gmail.com wrote:

<snip>
>>

>> Why not a zener or precision voltage IC?
>
>which one?

That was my question. ;-)

>
>> >with 0.1% resistors everywhere it matters.
>>
>> As long as the final result calibrates correctly, does it matter what
>> the tolerances are / were?
>
>There are 2 ways to get accuracy.
>1. adjust/calibrate

But adjustability sometimes brings with it lower precision (noise).

>2. build it accurately

Ok.

>I don't know which suits you better.

Whatever works best on the day. ;-)

>Long ago I used to use selected 5% resistors where 1% were wanted, but they don't stay as accurate as 1%ers.

Ok.

>
>
>> >Avoiding anything getting hot always helps keep the errors down,
>>
>> With 250uA FSD I can't anything much getting warm?
>
>no :)

;-)

>
>> >so anything that gets warm heatsink it, and operate parts way under their limits. Not that that will be a problem with this meter.
>>
>> Quite.
>>
>> So, precision reference, zener, resistor bridge?
>>
>> With a precision reference and zener my meter wouldn't start doing
>> anything till the measured voltage started to exceed 10V, even if I
>> was ramping from 0 to 15V (and that's fine for a 12V lead acid).
>
>If you had a perfect-knee perfectly accurate voltage zener then sure, but where will you find one of those?

Pass, hence the questions.

>
>> With a resistor bridge with a 1/3 split I should be able to get FSD at
>> my 15V but anything less wouldn't work (would it?) ... so it's going
>> to have to be either of the first two ... or something else?
>
>or resistors on the input of an opamp so it only turns on once 10v is reached

Ok, but you still need to hold the -Ve end of the meter *at* 10V as
the source goes between 10 and 15V? A LM4040DIZ-10.0 works like a
zener so that would probably do.

If I've read the datasheet properly the (10V) device requires a
minimum of 100uA and a maximum of 15mA.

http://www.ti.com/lit/ds/symlink/lm4040-n-q1.pdf

So, if we ignore the meter and assume a worst case battery voltage of
10.5V then we have .5V to get 100uA so that's 4k7 and giving us 1mA at
15V. We would also have 20500 ohms (20k series resistor and 500 ohms
of the meter) in parallel with the chip feed to help ensure it stays
over 100uA.

>> Basically, as long as I hold the negative of the meter at the same
>> voltage as the positive at anything 10V and below, the meter won't do
>> anything. Then as the +ve of the meter exceeds 10V and goes to 15V, a
>> meter set to read 0-5 volts should follow ok, giving me reasonable /
>> best possible accuracy across the working range of the battery.
>>
>> The only prerequisites are to ensure the -ve of the meter is held
>> exactly at +10V (from a supply ranging from 10.5 to 14.4V) and that
>> the meter current limiting resistor is correctly matched (to give the
>> 250uA required for FSD at 5V PD).
>>
>> As long as any reference would be suitably driving at anything above
>> 10V and not overdriven at 15V then we should be good to go?
>>
>> Cheers, T i m
>
>There's always protection to consider.

I think about all that would work (without affecting the accuracy /
functionality) is a very small series fuse and reverse biased diode
across the input to the meter circuit ?

Cheers, T i m

[T i m]

On Thu, 21 Sep 2017 18:11:32 -0500, Foxs Mercantile <jda...@att.net>
wrote:

But still, ohms law applies and with just a series resistor to limit
the current for FSD, zero on the meter (re-marked to display 10V)
won't reflect the required 10V.

For the meter to read zero there would have to be zero volts across it
and with just your series resistor that would only happen when the
supply voltage = zero.

So, a simple series resistor won't work for a meter reading from 10
(not 0) to 15V (but I'll be happy for you to explain it otherwise).
;-)

So, you need something to hold the -ve of the meter at 10V (and the
+ve of the meter to supply via a ~20k resistor) to allow the meter to
work in the range of 0 to 5V, marked to show 10 - 15V.

I hope that makes the requirement clearer. ;-)

Cheers, T i m

[mike]

OK,
Let's go back to engineering 101.
What's the VERY MOST important part of a project?
It's the SPECIFICATION.

Exactly, precisely, concisely, unambiguously, completely stated
requirements.
This does NOT change. If you change your mind and change the spec,
you start a new project.

How do you know when you're done? You write acceptance criteria.
Exactly how the user, in this case you, is going to test the unit
to determine if you met the spec? If it passes the written acceptance
criteria, you succeeded. If it passes the test criteria, but doesn't
work, the specification was the problem. You get paid for this project
and start a new one with new specifications.

I can't emphasize how important the specification is. It's the number
one cause of project failure. You can't design for criteria that
were not specified.
It's easy to say, "I know what I'm designing for me...I don't need
no stinkin' specification." You'd be wrong. This thread is what you
get. Lots of rocket scientists when you really need a boat.
The very last place you want to discover that your forgot some
important requirement is when the completed device is deployed.

Words like exactly, "as good as possible" have no place in a specification.

Specs are numbers and test methods. Specs include allowable
variations due to initial component tolerances, component age,
temperature, vibration, misuse.
An example of misuse is, "what happens when the user hooks it
up backwards." I learned about vibration when I carried a
voltmeter on my motorcycle. When I needed it, it failed.
I took it apart and it poured out like sand.

The engineer doesn't need to know WHY you want what you want.
That's not his job. HOWEVER...if you state your objectives,
why you want to do this, you might find the bevy of rocket scientists
have a much better way to accomplish your objectives. I can't
count the number of times a user wanted an complicated gizmo,
but his objective could be achieved with a much different
and much simpler approach by repartitioning the system.

End rant.

First question I'd ask is, "are you gonna' stare at the meter 24/7?"
If not, get a cheap digital panel meter off EBAY, put a push button
in series and be done with it. If you're not battery
powered, you can probably tolerate the load and don't need the
push button.

But, back to your original request.
Take your current meter.
Put a series resistor to set the max voltage to 5V.
Put a 10V "battery" in series.
The remaining question is, "how do you implement the 10V
battery?"
Can your measured voltage ever get below 10V?
If so, your problem just got a LOT more complicated.
See "misuse" above.

If it were me, I'd use a 10V zener diode. If you can find
an affordable temperature compensated one that is accurate
over the range of current from zero to the max indication
on your meter and the ambient temperature range, do that.
Characterize the calibration.
Fire up your CAD program and make a new meter face.
Since you're making a new face, any stable zener voltage works.
Your meter is as accurate as your care in making the new face.

Are we having fun yet?

[tabb...@gmail.com]

On Friday, 22 September 2017 00:23:14 UTC+1, T i m wrote:

> On Thu, 21 Sep 2017 14:37:35 -0700 (PDT), tabbypurr wrote:

> >> Why not a zener or precision voltage IC?
> >
> >which one?
>
> That was my question. ;-)
> >
> >> >with 0.1% resistors everywhere it matters.
> >>
> >> As long as the final result calibrates correctly, does it matter what
> >> the tolerances are / were?
> >
> >There are 2 ways to get accuracy.
> >1. adjust/calibrate
>
> But adjustability sometimes brings with it lower precision (noise).
>
> >2. build it accurately
>
> Ok.
>
> >I don't know which suits you better.
>
> Whatever works best on the day. ;-)

Both work fine. Which is easier.

> >Long ago I used to use selected 5% resistors where 1% were wanted, but they don't stay as accurate as 1%ers.
>
> Ok.
> >
> >
> >> >Avoiding anything getting hot always helps keep the errors down,
> >>
> >> With 250uA FSD I can't anything much getting warm?
> >
> >no :)
>
> ;-)
> >
> >> >so anything that gets warm heatsink it, and operate parts way under their limits. Not that that will be a problem with this meter.
> >>
> >> Quite.
> >>
> >> So, precision reference, zener, resistor bridge?
> >>
> >> With a precision reference and zener my meter wouldn't start doing
> >> anything till the measured voltage started to exceed 10V, even if I
> >> was ramping from 0 to 15V (and that's fine for a 12V lead acid).
> >
> >If you had a perfect-knee perfectly accurate voltage zener then sure, but where will you find one of those?
>
> Pass, hence the questions.

http://www.datasheetcatalog.com/datasheets_pdf/B/Z/X/5/BZX55C10.shtml
shows that the breakdown voltage is below nominal at 0-250uA, and resistance far higher than at 5mA specified.
With Ir 0.1uA @ 7.5v you could use one, but you'd need to accept a yet unknown amount of nonlinearity in the scale. Far easier to just design something more accurate.

> >> With a resistor bridge with a 1/3 split I should be able to get FSD at
> >> my 15V but anything less wouldn't work (would it?) ... so it's going
> >> to have to be either of the first two ... or something else?
> >
> >or resistors on the input of an opamp so it only turns on once 10v is reached
>
> Ok, but you still need to hold the -Ve end of the meter *at* 10V as
> the source goes between 10 and 15V? A LM4040DIZ-10.0 works like a
> zener so that would probably do.

The meter can just go on the opamp output, the offbiasing can be done on the input side.

> If I've read the datasheet properly the (10V) device requires a
> minimum of 100uA and a maximum of 15mA.
>
> http://www.ti.com/lit/ds/symlink/lm4040-n-q1.pdf
>
> So, if we ignore the meter and assume a worst case battery voltage of
> 10.5V then we have .5V to get 100uA so that's 4k7 and giving us 1mA at
> 15V. We would also have 20500 ohms (20k series resistor and 500 ohms
> of the meter) in parallel with the chip feed to help ensure it stays
> over 100uA.

or don't use it.

> >> Basically, as long as I hold the negative of the meter at the same
> >> voltage as the positive at anything 10V and below, the meter won't do
> >> anything. Then as the +ve of the meter exceeds 10V and goes to 15V, a
> >> meter set to read 0-5 volts should follow ok, giving me reasonable /
> >> best possible accuracy across the working range of the battery.
> >>
> >> The only prerequisites are to ensure the -ve of the meter is held
> >> exactly at +10V (from a supply ranging from 10.5 to 14.4V) and that
> >> the meter current limiting resistor is correctly matched (to give the
> >> 250uA required for FSD at 5V PD).
> >>
> >> As long as any reference would be suitably driving at anything above
> >> 10V and not overdriven at 15V then we should be good to go?
> >>
> >> Cheers, T i m
> >
> >There's always protection to consider.
>
> I think about all that would work (without affecting the accuracy /
> functionality) is a very small series fuse and reverse biased diode
> across the input to the meter circuit ?
>
> Cheers, T i m

Enough resistance on the opamp input buys a lot of protection, a reverse diode across it buys more. Power line diode gets you more. You could add diodes to avoid one power section discharging while the other stays up.

Re running the meter off the voltage it's measuring, you could avoid some issues by having a 2nd opamp section pump up its supply rail voltage. Ultimate rail accuracy isn't required, the one opamp could work as both oscillator & rail voltage comparator.

Or easier, just cut Vin in half and run the opamp off the 12v battery.


NT

[tabb...@gmail.com]

He's making a one off for use at home, not commissioning a new design of military hardware.

> First question I'd ask is, "are you gonna' stare at the meter 24/7?"
> If not, get a cheap digital panel meter off EBAY, put a push button
> in series and be done with it.

not compliant with the spec


NT

[Foxs Mercantile]

On 9/21/2017 6:35 PM, T i m wrote:
> But still, ohms law applies and with just a series resistor to limit
> the current for FSD, zero on the meter (re-marked to display 10V)
> won't reflect the required 10V.

Now, ya see, I missed that.
you want a 10-15 volt meter, not a 0-15 volt meter.

The simplest is a 20K resistor in series with a 10v zener diode.
Use a low current Zener so the knee works right at low values.
Something like this:
<http://www.mouser.com/ds/2/395/BZX55C2V0%20SERIES_D1610-1099630.pdf>

Depending on how non-linear the low end is, you can print a custom
dial scale that has been corrected for "this voltage is this mark."
<http://www.tonnesoftware.com/meter.html>
Although this one looks to be linear down to 10 uA or better.
10 uA would be approximately 0.2 volts above the Zener voltage.
or 10.2 volts on a 10-15 volt scale.

[tabb...@gmail.com]

simple but time consuming producing a new dial paper


NT

[Jeff Liebermann]

On Thu, 21 Sep 2017 20:59:03 -0700 (PDT), tabb...@gmail.com wrote:

>simple but time consuming producing a new dial paper
>NT

"The Windows programs for drawing analog meter scales"
<http://www.tonnesoftware.com/meter.html>
$35 for the full version.

--
Jeff Liebermann je...@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

[Foxs Mercantile]

On 9/21/2017 10:59 PM, tabb...@gmail.com wrote:
> simple but time consuming producing a new dial paper

With the low current Zener, it doesn't look like that will be
necessary.

[mike]

On 9/21/2017 8:13 PM, tabb...@gmail.com wrote:

>> End rant.
>
> He's making a one off for use at home, not commissioning a new design of military hardware.
>

The task is the same no matter who the customer.
The more steps you skip, the more problems you have.
The more you think like a project manager, the better
managed your home projects.
Once you acquire the habit, you'll find that it adds
little to the time in the beginning and greatly simplifies
the end.
And it wastes far less time for rocket scientists in residence.

[mike]

On 9/21/2017 9:05 PM, Jeff Liebermann wrote:
> On Thu, 21 Sep 2017 20:59:03 -0700 (PDT), tabb...@gmail.com wrote:
>
>> simple but time consuming producing a new dial paper
>> NT
>
> "The Windows programs for drawing analog meter scales"
> <http://www.tonnesoftware.com/meter.html>
> $35 for the full version.
>

I haven't seen the system requirements, but I'd wager that
the precision measurement desired is never entered into
any other system/program. The desired functionality
could be obtained by a Sharpie mark or two on the face of the meter.

One of the difficult engineering skills to acquire is the ability
to match the solution to the actual problem. Overkill is rampant.
It's what you do when you don't understand what you need, but you
start building it anyway.

If you need precision, you don't want an analog meter.

[Foxs Mercantile]

On 9/21/2017 11:42 PM, mike wrote:
> I haven't seen the system requirements, but I'd wager that
> the precision measurement desired is never entered into
> any other system/program.  The desired functionality
> could be obtained by a Sharpie mark or two on the face of the meter.
>
> One of the difficult engineering skills to acquire is the ability
> to match the solution to the actual problem.  Overkill is rampant.
> It's what you do when you don't understand what you need, but you
> start building it anyway.
>
> If you need precision, you don't want an analog meter.

The meter program is capable of 100 "placed" I.e. non-linear tick
marks over the scale.
Much like some of the expensive lab grade equipment built with hand
lettered scales.

I am reminded of a story in Model Railroader back in the early '50s.
One of their staff members was in South Korea during the Korean war.
He was bound and determined to build himself a small table railroad.
He'd spent several weeks making sure every measurement was right and
the corners were square.

Finally, the big day arrived. He placed it in the corner of the
living room. Much to his horror, he had a triangular gap along one
side.

He asked his Korean attache how could he have possible had that
happen. The answer was simple, "Korean house not always square."

"But, your furniture and tables fit perfectly!"

"Korean tables and furniture not always square either."

[T i m]

On Thu, 21 Sep 2017 17:21:23 -0700, mike <ham...@netzero.net> wrote:

<snip read and well meaning stuff for brevity>

>
>Words like exactly, "as good as possible" have no place in a specification.

I guess that depends on your background, viewpoint and goals mike?

e.g. I knew what I wanted, had a rough idea how it could be done and
just thought I run it past the panel for some real world *design*
feedback (I'm very much a dabbling designer but long term builder /
support guy etc).

>
>Specs are numbers and test methods. Specs include allowable
>variations due to initial component tolerances, component age,
>temperature, vibration, misuse.

Understood. I felt I laid out the relevant spec on my initial post, or
at least as far as someone who understood humans were concerned. ;-)

"Not a 'repair' question as such but I was thinking of using an old
skool 250uA FSD moving coil panel meter to read the 10 to 15V DC range
(a lead acid battery charge / discharge indicator), next to a 0-30A
Ammeter (inc shunt etc).

I was thinking of using either a voltage reference such as
LM4040DIZ-10.0 or could I get away with a simple zener (BZX55C?) with
associated resistor(s) please?"

So that was the need, the specs as far as I knew them and a potential
solution to help people 'get' what I was considering. What I didn't do
was signpost the fact that I was actually using a 0-5V panel meter to
display the 10-15V range. ;-(

The 'human' bit: A small (key fact) analogue (key fact) panel meter
connected to a lead acid battery (key fact) needs reading and those
readings interpreting and so ITRW, we are 'most likely' to not require
much in the way of precision or we wouldn't be using those things in
the first place. ;-)

So, what I was hoping I was saying is 'Could someone confirm or
recommend that 'for the (reading_between_the_lines) reasonably 'soft'
requirements, what is likely to offer *a* (there may be many equally
competent) real word / human solution?

So, anyone who knows their stuff and who understands people *might*
have already answered the question on 'how much money', 'how much
effort', 'how complex / simple' would any solution likely to be in
their heads. <shrug>

>An example of misuse is, "what happens when the user hooks it
>up backwards."

Understood, but unbeknown to anyone here, that will all be catered for
*in my case* by the use of polarised connectors that could never end
up reverse connected elsewhere.

>I learned about vibration when I carried a
>voltmeter on my motorcycle. When I needed it, it failed.
>I took it apart and it poured out like sand.

Again, understood. In this instance the only clue I gave to the
potential size of the battery was the scale of the ammeter (0-30A) and
that scale was because it is to be used on an electric outboard motor
that has a maximum current of 30A and will in fact be running from 3 x
31Ah batteries joined in parallel. Because it's in a boat the panel
meters will have a very comfortable ride. ;-)

>
>The engineer doesn't need to know WHY you want what you want.

Quite. ;-)

>That's not his job. HOWEVER...if you state your objectives,
>why you want to do this, you might find the bevy of rocket scientists
>have a much better way to accomplish your objectives.

Again, how would knowing what I wanted to use it for, more than I
already specified help anyone actually resolve the question asked? I'm
not saying they couldn't then rule out or add 'other things' but
nothing offered so far has affected the physical outcome of this
particular project in a physical sense.

That said, I could have been using a large and expensive panel meter,
graduated in .2v increments, rather than one only marked in whole volt
increments <g>. But the reading of the meter is my problem, making it
read 10-15V is the bit I needed help on (please). ;-)

As an aside here ... I ran a Telephone Help Desk for over 5 years and
feel I became fairly good at 'gauging' people. If their opening gambit
was 'Your beige box isn't working' I knew to deal with them at a
different level (initially at least) compared with someone who said
'Our X.25 PAD No3 doesn't seem to be bringing up a Link LED?'.

Just the same as the PC user reporting that their 'screen is blank'
when it in fact has a flashing cursor in the top left corner. It's not
what they expected or wanted but it's far from 'blank'. ;-)

>I can't
>count the number of times a user wanted an complicated gizmo,
>but his objective could be achieved with a much different
>and much simpler approach by repartitioning the system.

Oh, absolutely and you / anyone (who hasn't already dealt with me on
here) wouldn't have any idea what I do or do not know and may have
already considered. My solution potentially required 2 components
(resistor and zener) so it couldn't get much simpler so it's just a
matter of selecting the 'best' (all things considered) types of those
components, if they are appropriate.
>
>End rant.

Points noted and generally agreed mike. ;-)

>
>First question I'd ask is, "are you gonna' stare at the meter 24/7?"

No, it's something I'll glance at either now and again or specifically
when I change something and need to see the consequences (like
changing 'speed' on my electric outboard.

>If not, get a cheap digital panel meter off EBAY,

Already Done.

>put a push button
>in series and be done with it.

No need mike. It will likely only be in circuit when required
(testing) and the current drawn when it is will be insignificant
relative to the other loads and battery capacity (93Ah).

> If you're not battery
>powered, you can probably tolerate the load and don't need the
>push button.

Check.

>
>But, back to your original request.

Cool. ;-)

>Take your current meter.

Check.

>Put a series resistor to set the max voltage to 5V.

Check (~20k?)

>Put a 10V "battery" in series.

Bench PSU + DMM, check.

>The remaining question is, "how do you implement the 10V
>battery?"

Check.

>Can your measured voltage ever get below 10V?

Hopefully not. It will only ever be 10.5V or higher or 14.4V or lower
... or disconnected completely (points already considered etc).

>If so, your problem just got a LOT more complicated.
>See "misuse" above.

Quite. ;-)

>
>If it were me, I'd use a 10V zener diode. If you can find
>an affordable temperature compensated one that is accurate
>over the range of current from zero to the max indication
>on your meter and the ambient temperature range, do that.

Check. I did suggested a BZX55C in my OP but not being a designer I
would have no real idea where to start re finding the 'right'
component (by that I mean there is probably a 'go-to' device that
'most people' would typically use under these circumstances, like they
might a 7805 or 555) ;-)

>Characterize the calibration.

Ok?

>Fire up your CAD program and make a new meter face.

Unlikely (see all the above). ;-)

>Since you're making a new face, any stable zener voltage works.

Chances are I'm not so ...

>Your meter is as accurate as your care in making the new face.

Quite ... and with simple 1V increments over a 10-15V range should
give me a better 'feel' of the voltage than trying to read a similarly
sized meter graduated in .2v increments in the 0-15V range all whilst
going along a river in a small folding boat. ;-)

>
>Are we having fun yet?

Very much so, thanks mike. ;-)

Cheers, T i m

[T i m]

On Thu, 21 Sep 2017 21:41:44 -0700, mike <ham...@netzero.net> wrote:

>On 9/21/2017 8:13 PM, tabb...@gmail.com wrote:
>
>>> End rant.
>>
>> He's making a one off for use at home, not commissioning a new design of military hardware.
>>
>
>The task is the same no matter who the customer.

It can and maybe should be, in an ideal world. I say that because I
hate having to deal with 'fools who rush in' or those expecting me to
make technical / choice decisions on their behalf (like getting them
some 'size 10, brown, lace up leather shoes'). ;-(

>The more steps you skip, the more problems you have.

'May have'.

>The more you think like a project manager, the better
>managed your home projects.

Agreed. I like to understand as much about something as I can.
However, my ability to do so, time available to do so and the
pertinence of doing so don't always allow me to do so or to do so to
the level I would like. Sometimes you just have to get stuck in. ;-)

>Once you acquire the habit, you'll find that it adds
>little to the time in the beginning and greatly simplifies
>the end.

Agreed 100%.

>And it wastes far less time for rocket scientists in residence.

Quite ... however, this is a 'discussion' group and so does have an
element of that, along with the hard science / fact stuff.

Part of the 'human' bit is not making a post (especially the initial
one) too long (I fail most of the time) as it will put people off
reading it and if there were any further points that need refining
they could be done with a simple Q&A.

e.g. I initially outlined the fairly close detail at a level relevant
to my needs and understanding (IMHO anyway). I didn't expand on what I
was going to actually use it for because it didn't really matter to
the question. I left it open to the reader to ask any supplementary
questions as they felt relevant. ;-)

Cheers, T i m

[Dave M]

Wow, what a long thread about such a simple problem. Here's a web page
describing exactly the circuit you need for an expanded scale voltmeter
(http://sound.whsites.net/articles/meters.htm). Scroll down to Apragraph
5.0 to get to the expanded voltmeter discussion. It uses a 1ma movement for
the discussion, so you'll have to calculate the resistor values to fit your
meter. It uses a 10V zener and one more resistor.

Another method that I've used in the past to make an expanded scale meter is
to suppress the mechanical zero. You still have (in your case) a 0-15V
meter, but the needle doesn't get up to the "0" mark until 10V is felt
across the circuit. After that, the meter responds normally, up to 15V full
scale. So,with this approach, you don't need any extra circuitry to make an
expanded scale meter.
This means that you'll have to open the meter and (CAREFULLY!!) move the
mechanical zero tang. First, apply 10V to the circuit. Then, turn the
mechanical zero tang so that the pointer goes down to the "0" mark.
Reassemble the meter movement, and you're all done.

To make this approach work, your meter movement must have enough mechanical
adjustment range to suppress the pointer enough so that it needs 10V across
the circuit to make the meter read "0" V. Typically, D'Arsonval movements
have about 20-30% range on the mechancial zero adjustment, so yours might be
a long stretch for this method. Never know till you try!!
Dave M

[Dave M]

I forgot to mention that for the zener solution, you really need to select
as low power a zener that you can find, such as 400mW or lower.
The 1N4104 is a 10V, 250mW zener, with a test current of 1uA, and a max
current of 25ma, well within your requirements. In stock at Mouser.com.

Cheers,
Dave M

[Jeff Liebermann]

In college (about 1969), I build an "electronic slide rule" in the
form of an analog computer inside a brief case. Inside, it was two
log amps, a multiplier, range switch, and an anti-log amp. Input was
via two Bournes Helipot 10 turn pots (linearity 0.1%). Output was on
a mirrored scale analog panel meter about 6" in width. Accuracy was 3
decimal places in the lower part of the meter, and 2 decimal places
over the rest of the scale. However, I had to use temperature
compensation tables for the pots, multipliers, and amps to obtain that
level of accuracy.

What level of precision were you expecting?

[Ian Field]

"T i m" <ne...@spaced.me.uk> wrote in message

news:e898schb0qaru15b3...@4ax.com...

> On Thu, 21 Sep 2017 21:43:31 +0100, "Ian Field"
> <gangprobi...@virginmedia.com> wrote:
>
>>
>>
>>"T i m" <ne...@spaced.me.uk> wrote in message
>>news:vas7scdd02ptrj9ig...@4ax.com...
>>> Hi All,
>>>
>>> Not a 'repair' question as such but I was thinking of using an old
>>> skool 250uA FSD moving coil panel meter to read the 10 to 15V DC range
>>> (a lead acid battery charge / discharge indicator), next to a 0-30A
>>> Ammeter (inc shunt etc).
>>>
>>> I was thinking of using either a voltage reference such as
>>> LM4040DIZ-10.0 or could I get away with a simple zener (BZX55C?) with
>>> associated resistor(s) please?
>>
>>Most people just use a Zener and maybe a few normal (or SB) forward volt
>>drops to trim the offset.
>
> Ok.
>>
>>I was thinking the TL431 programmable Zener could be calibrated to exactly
>>what you need - but it needs a minimum current through it (possibly 1mA or
>>so).
>
> Well that's ok?
>>
>>You can probably incorporate a TL431 into some kind of bridge circuit with
>>the movement spanning the 2 arms, but it will almost certainly need some
>>kind of power supply.
>
> Because the reference voltage will be too close to the minimum voltage
> expected Ian?

The TL431 is actually a comparator with its own built in 2.5V reference.

In closed loop operation, that's as low as the cathode goes - you can
saturate it in open loop operation, but "Vce-sat" is around 2V.

[whit3rd]

On Thursday, September 21, 2017 at 10:51:33 AM UTC-7, T i m wrote:
> Hi All,
>
> Not a 'repair' question as such but I was thinking of using an old
> skool 250uA FSD moving coil panel meter to read the 10 to 15V DC range
> (a lead acid battery charge / discharge indicator), next to a 0-30A
> Ammeter (inc shunt etc).
>
> I was thinking of using either a voltage reference such as
> LM4040DIZ-10.0 or could I get away with a simple zener (BZX55C?) with
> associated resistor(s) please?

You can use any zener under 10V to do a 10V zero offset (just divide down the
input voltage until that 10V input matches the zener knee). So, with an
arbitrary zener (I'd use a TL431 as a 2.5V reference, they're convenient
and accurate) and a trimmer acting as voltage divider, it just remains
to make a range-setting resistor in series with the meter.

The important thing, is to doodle up a really nifty scale for your meter
that indicates 10V to 15V, with clear markings, along the pointer's arc,
maybe with a bunch of subdivision marks. Last time I did that, it took
a plotter and a bit of custom software to draw the arcs and labels,
then some fiddling with the faceplate of the meter to
affix it (warning: you need to worry about laser-print ink, paper, glue compatibility).

[tabb...@gmail.com]

On Friday, 22 September 2017 05:05:35 UTC+1, Jeff Liebermann wrote:

> On Thu, 21 Sep 2017 20:59:03 -0700 (PDT), tabbypurr wrote:

> >simple but time consuming producing a new dial paper
>

> "The Windows programs for drawing analog meter scales"
> <http://www.tonnesoftware.com/meter.html>
> $35 for the full version.

fails to solve the issue AND adds pointless cost.


NT

[tabb...@gmail.com]

On Friday, 22 September 2017 05:22:46 UTC+1, Foxs Mercantile wrote:

> On 9/21/2017 10:59 PM, tabbypurr wrote:
> > simple but time consuming producing a new dial paper
>
> With the low current Zener, it doesn't look like that will be
> necessary.

Oh, you've found a zener that's 10v +/- 1%?


NT

[tabb...@gmail.com]

On Friday, 22 September 2017 05:42:36 UTC+1, mike wrote:

> On 9/21/2017 8:13 PM, tabbypurr wrote:

> >> End rant.
> >
> > He's making a one off for use at home, not commissioning a new design of military hardware.
> >
>
> The task is the same no matter who the customer.

Rubbish. For a large project all the specs have been calculated to meet the requirements, the contract gone through by a team of lawyers, and the task of the lec eng is to produce a circuit that meets all the given specs under all specified conditions.

For a one off at home the requirements are often not pinned down firmly, as in this case. There are no lawyers, no contract, and the elec eng's job is to trade off cost & complexity versus specs to find what turns out to be the most agreeable compromise for the client, oneself. In most cases this is a device that is much less tightly specced.


NT

[tabb...@gmail.com]

On Friday, 22 September 2017 05:43:10 UTC+1, mike wrote:
> On 9/21/2017 9:05 PM, Jeff Liebermann wrote:

> > On Thu, 21 Sep 2017 20:59:03 -0700 (PDT), tabbypurr wrote:
> >
> >> simple but time consuming producing a new dial paper
> >

> > "The Windows programs for drawing analog meter scales"
> > <http://www.tonnesoftware.com/meter.html>
> > $35 for the full version.
> >
> I haven't seen the system requirements, but I'd wager that
> the precision measurement desired is never entered into
> any other system/program. The desired functionality
> could be obtained by a Sharpie mark or two on the face of the meter.

Indeed. This can be done in such a way to make it tidy. But the use of a zener means the scale will have to be calibrated, which is a waste of time, especially to get a nice neat result.

> One of the difficult engineering skills to acquire is the ability
> to match the solution to the actual problem. Overkill is rampant.
> It's what you do when you don't understand what you need, but you
> start building it anyway.
>
> If you need precision, you don't want an analog meter.

I see it's foolish claims day today.


NT

[tabb...@gmail.com]

> On Thu, 21 Sep 2017 17:21:23 -0700, mike <ham...@netzero.net> wrote:

>Words like exactly, "as good as possible" have no place in a specification.

a common enough phrase for an in-house spec. Such phrases should be cleared up later of course, once the specifier knows what the engineers can actually do and what the options cost.

> >Specs are numbers and test methods. Specs include allowable
> >variations due to initial component tolerances, component age,
> >temperature, vibration, misuse.

Some do. Some are 'how can you get it as cheap as possible and still have it sell & not destroy our repuation.' Some are 'we want to beat what's out there, how good could you get it?' Some are 'we want it to do this.' etc etc etc.

> The 'human' bit: A small (key fact) analogue (key fact) panel meter
> connected to a lead acid battery (key fact) needs reading and those
> readings interpreting and so ITRW, we are 'most likely' to not require
> much in the way of precision or we wouldn't be using those things in
> the first place. ;-)

1% is doable for some analogue meters. That's enough precision for a lot of tasks.


NT

[amdx]

On 9/21/2017 3:18 PM, pf...@aol.com wrote:
> Like this?
>
> http://thumbs.ebaystatic.com/images/g/XAMAAOSw-0xYRaJT/s-l225.jpg
>
> Right off the shelf.
>
> Several others from similar sources with a table of resistance values to make the desired range (in volts) and either blank faces or sequential numeric faces. Sorry for pointing to an Euro-meter, but just that they are out there.
>
> Had I known, the "meter man" at Kutztown would have had exactly what you want in the size you want it, new or used. He has hundreds. Now you have to wait until spring.
>
> Peter Wieck
> Melrose Park, PA
>

I don't know if you are asking for an expanded scale meter, were the
0V line is 10V and FS line is 15 volts, so the whole range is only 5
volts. That's what I gleaned.
First you need to put a series 60K ohm resistor to make the meter read
FS with 15V applied. (for accuracy you should subtract the resistance of
the meter from the 60K ohms)
Then you need to add a circuit shown in these sites.

> http://www.radcomms.net/Expanded_Scale_Voltmeter.html

http://www.thebackshed.com/windmill/articles/ExpandedScaleVoltMeter.asp

> http://www.zbasic.net/forum/about1423.html

This last one will probably need a 10K resistor tied from one OUT to
the other OUT. Just so there is enough current to make the regulators
work properly.

Here's a youtube, I didn't watch it all, but you might learn something.

> https://www.youtube.com/watch?v=-z_j4MclCTk

Mikek

[bruce2...@gmail.com]

mike wrote:
> On 9/21/2017 8:13 PM, tabbypurr wrote:

> >> End rant.
> >
> > He's making a one off for use at home, not commissioning a new design of military hardware.
>

> The task is the same no matter who the customer.

At least the extra option should be there.

[Jeff Liebermann]

The consensus seems to be to use an expanded scale analog voltmeter.
If the OP is going to build such a thing, he needs a new scale on his
analog meter. That can be done with the free version of the software.
(I just checked and it will do an expanded scale).

I usually ignore one line pontifications, but I'm curious. Why does a
new meter scale fail to solve the issue? Ummm... what issue?

[Jeff Liebermann]

The meter isn't even that accurate. You're not going to be able to
read the meter to +/- 0.1V. Also, I coulnd't find where anyone
specified the accuracy required.

If you really want 1% accuracy, maybe a 10.0V precision reference,
such as the TI REF102:
<http://www.ti.com/lit/ds/symlink/ref102.pdf>
However, there's a problem. The IC requires a V+ of 11.4-36VDC which
will probably require a battery or other power supply.

Hmmm... I sorta blundered across this idea:
<http://www.qsl.net/kh6grt/page4/expscale/expscale.htm>
It doesn't expand the entire meter scale from 10-15VDC.
Instead, it compresses 0-10V into a small part of the meter scale, and
expands the 10-15V over the rest of the scale. However, it requires
+12v and -12v power.

[tabb...@gmail.com]

On Saturday, 23 September 2017 17:35:36 UTC+1, Jeff Liebermann wrote:

> On Fri, 22 Sep 2017 15:38:26 -0700 (PDT), tabbypurr wrote:
> >On Friday, 22 September 2017 05:05:35 UTC+1, Jeff Liebermann wrote:
> >> On Thu, 21 Sep 2017 20:59:03 -0700 (PDT), tabbypurr wrote:
> >
> >> >simple but time consuming producing a new dial paper
> >>
> >> "The Windows programs for drawing analog meter scales"
> >> <http://www.tonnesoftware.com/meter.html>
> >> $35 for the full version.
>
> >fails to solve the issue AND adds pointless cost.
> >NT
>
> The consensus seems to be to use an expanded scale analog voltmeter.

that's part of the spec

> If the OP is going to build such a thing, he needs a new scale on his
> analog meter.

yes

> That can be done with the free version of the software.
> (I just checked and it will do an expanded scale).

it can

> I usually ignore one line pontifications, but I'm curious. Why does a
> new meter scale fail to solve the issue? Ummm... what issue?

The issue with using an inaccurate zener is that the meter scale most likely won't run from 10.0v, and will thus need calibrating over its scale. That also means a new scale with marks in new places - a pita to do. Far easier to knock off 10v exactly then just change the numbers on the scale.


NT

[tabb...@gmail.com]

On Saturday, 23 September 2017 17:52:35 UTC+1, Jeff Liebermann wrote:

> On Fri, 22 Sep 2017 15:39:14 -0700 (PDT), tabbypurr wrote:
> >On Friday, 22 September 2017 05:22:46 UTC+1, Foxs Mercantile wrote:
> >> On 9/21/2017 10:59 PM, tabbypurr wrote:

> >> > simple but time consuming producing a new dial paper
> >>
> >> With the low current Zener, it doesn't look like that will be
> >> necessary.
>
> >Oh, you've found a zener that's 10v +/- 1%?
> >NT
>
> The meter isn't even that accurate.

We don't know the meter's accuracy. 1% is common enough, but it might be worse. (For all we know it might even be a moving iron thing.)

> You're not going to be able to
> read the meter to +/- 0.1V.

Even my most rock bottom multimeters, under $3 each new, are easy to read to 1%. Decent meters do much better.

> Also, I coulnd't find where anyone
> specified the accuracy required.
>
> If you really want 1% accuracy, maybe a 10.0V precision reference,
> such as the TI REF102:
> <http://www.ti.com/lit/ds/symlink/ref102.pdf>
> However, there's a problem. The IC requires a V+ of 11.4-36VDC which
> will probably require a battery or other power supply.
>
> Hmmm... I sorta blundered across this idea:
> <http://www.qsl.net/kh6grt/page4/expscale/expscale.htm>
> It doesn't expand the entire meter scale from 10-15VDC.
> Instead, it compresses 0-10V into a small part of the meter scale, and
> expands the 10-15V over the rest of the scale. However, it requires
> +12v and -12v power.

Nice idea, though I don't expect the op needs it. I'm sure one could design a circuit that can use opamps that are happy on the 10-15v and halve V_in.


NT

[mike]

Metrology is an interesting topic.

When someone asks me how to measure something, my first thought is,
"exactly how will the answer to your question improve YOUR life tomorrow?"
Sometimes I actually say it.
The "jolt" to their thought process forces them to express what they're
trying to accomplish...or admit to themselves that they have no idea.
I find that almost all questions end up in the "won't affect them at all"
category and I can dispense with trying to teach them something they'll
never use. If the answer won't change the future, don't bother with the
question.

It's similar to pointless social interaction.
People you don't know ask questions, like "how are you?"
They think they know what you'll say.
Respond with something like, "I have this rash on my dick and
think I need to find a different street corner... got any ideas?"
to see how much they really care about how you are. ;-)
But I digress...

If I had an electric boat, the ability to measure voltage to six
decimal places would not be my priority.
What I want to know is, "can I get back to the dock?"
You can do that more accurately with a sharpie and nonlinear
meter than you can react to a change in wind direction.

The river current and wind and the shape of the curve of distance covered vs
velocity and temperature would be far more important variables to me.
Amp hours consumed would be important.

Battery voltage that would be strongly impacted by whether I'm going
up or down stream at what net velocity is a poor substitute.

An integrating current meter, a water speed speedometer and a GPS would
much more interesting. If you knew the water flow direction and speed,
you probably wouldn't have to measure anything else other than
what you can get from the GPS. Drifting downstream for a minute
would give you those two numbers.

But, since you're dead set on measuring voltage...
The basic problem is that the voltage you want to measure is
the biggest voltage you have to power your circuit.
That's why I split my original post into the concept (battery)
and the implementation (zener).
A series zener needs to draw ZERO current at 10V and still
be 10V at MUCH higher current ratio. You will have to live
with some amount of nonlinearity.

You could stack an AA-cell or two on top of your battery system
and alleviate that problem.

I maintain that a digital voltmeter is the best option.
They're dirt cheap on EBAY.

All the above is based on an extreme extrapolation of
very limited information. That's why the system spec is so important.

[tabb...@gmail.com]

On Sunday, 24 September 2017 12:06:45 UTC+1, mike wrote:

> A series zener needs to draw ZERO current at 10V and still
> be 10V at MUCH higher current ratio. You will have to live
> with some amount of nonlinearity.

In fact you could re-zero the meter so it reads 10v spot on. Add an extra mark for where it points when unpowered so you can see if it ever decalibrates.

> You could stack an AA-cell or two on top of your battery system
> and alleviate that problem.

yuck. 2 resistors beats 2 cells any day.

> I maintain that a digital voltmeter is the best option.
> They're dirt cheap on EBAY.

but noncompliant with the basic spec


NT

[mike]

On 9/24/2017 6:31 AM, tabb...@gmail.com wrote:
> On Sunday, 24 September 2017 12:06:45 UTC+1, mike wrote:
>
>> A series zener needs to draw ZERO current at 10V and still
>> be 10V at MUCH higher current ratio. You will have to live
>> with some amount of nonlinearity.
>
> In fact you could re-zero the meter so it reads 10v spot on. Add an extra mark for where it points when unpowered so you can see if it ever decalibrates.
>

Yes, you can , but it's still nonlinear.
It's a matter of how much nonlinearity you can tolerate.
It's my assertion that the measured quantity is so crude that you don't
need to worry much about anything. Just learn what it reads and calibrate
your expectations to match.

>
>> You could stack an AA-cell or two on top of your battery system
>> and alleviate that problem.
>
> yuck. 2 resistors beats 2 cells any day.

Can't argue with that assertion, but, if you can
provide a circuit that is precisely linear and accurate over the WHOLE
range using
two resistors, I'd like to see it. I'm betting it's more than two
resistors.

>
>> I maintain that a digital voltmeter is the best option.
>> They're dirt cheap on EBAY.
>
> but noncompliant with the basic spec

Yep, that's a problem with any customer expectation.
I try to adjust expectations to meet the ACTUAL requirements
before delving into complex solutions to implement a nonsense
requirement.
Yes, the customer is always right. My job is to alert him
to the consequences of his demands.

Another non-complient solution is to use the two resistors
and a zener
to make a 9V-15V meter and hope that most of the nonlinearity
of the two resistor solution happens in the 9-10V range.
That way, the characteristics of the zener knee are important
over a 5:1 current range instead of an infinite ratio.
I'm far too lazy to do the math on how nonlinear it will be.

There's also an elephant in the room.
Unless you start with a meter that happens to have scale
markings ranging from 10 to 15, you're gonna need a new scale
no matter what you do with the electronics. If you're not gonna
do that, might as well just mark calibrations on the face with a
sharpie.
Makes no sense to go to great lengths in one part of the project
only to create something that still doesn't do what you want.
>
>
> NT
>

[tabb...@gmail.com]

On Monday, 25 September 2017 01:53:43 UTC+1, mike wrote:

> On 9/24/2017 6:31 AM, tabbypurr wrote:
> > On Sunday, 24 September 2017 12:06:45 UTC+1, mike wrote:
> >
> >> A series zener needs to draw ZERO current at 10V and still
> >> be 10V at MUCH higher current ratio. You will have to live
> >> with some amount of nonlinearity.
> >
> > In fact you could re-zero the meter so it reads 10v spot on. Add an extra mark for where it points when unpowered so you can see if it ever decalibrates.
> >
> Yes, you can , but it's still nonlinear.
> It's a matter of how much nonlinearity you can tolerate.
> It's my assertion that the measured quantity is so crude that you don't
> need to worry much about anything. Just learn what it reads and calibrate
> your expectations to match.
> >
> >> You could stack an AA-cell or two on top of your battery system
> >> and alleviate that problem.
> >
> > yuck. 2 resistors beats 2 cells any day.
> Can't argue with that assertion, but, if you can
> provide a circuit that is precisely linear and accurate over the WHOLE
> range using
> two resistors, I'd like to see it. I'm betting it's more than two
> resistors.

What I meant is divide the input to your opamp so it's away from the rails and you don't need any rail boosting.


NT

[T i m]

On Sun, 24 Sep 2017 17:52:51 -0700, mike <ham...@netzero.net> wrote:

>On 9/24/2017 6:31 AM, tabb...@gmail.com wrote:
>> On Sunday, 24 September 2017 12:06:45 UTC+1, mike wrote:
>>
>>> A series zener needs to draw ZERO current at 10V and still
>>> be 10V at MUCH higher current ratio. You will have to live
>>> with some amount of nonlinearity.
>>
>> In fact you could re-zero the meter so it reads 10v spot on. Add an extra mark for where it points when unpowered so you can see if it ever decalibrates.
>>
>Yes, you can , but it's still nonlinear.
>It's a matter of how much nonlinearity you can tolerate.
>It's my assertion that the measured quantity is so crude that you don't
>need to worry much about anything. Just learn what it reads and calibrate
>your expectations to match.

(I've been a bit preoccupied elsewhere but still following the thread
with interest and will reply individually when I get the chance)

That was part I had already considered in that:
The meter is fairly small. The meter is glued shut. The meter is only
marked in whole volts.

So, as you say, afa this experiment is concerned it was a cheap
toe-dipping into the concept of meter range enhancement, partly
because I think analogue display still have some merit and partly for
the S&G's. ;-)
>>

<snip>

>I try to adjust expectations to meet the ACTUAL requirements
>before delving into complex solutions to implement a nonsense
>requirement.

In this case the actual requirements were fairly simplistic as it was
an experiment. A feasibility study if you will. As you say, a DMM can
be bought for next to nothing (and I have several and one I'm already
using) but that wasn't the (entire) point.

>Yes, the customer is always right. My job is to alert him
>to the consequences of his demands.

;-)

>
>Another non-complient solution is to use the two resistors
>and a zener
>to make a 9V-15V meter and hope that most of the nonlinearity
>of the two resistor solution happens in the 9-10V range.

With a very small buck converter you could get a higher voltage supply
to ensure a voltage reference of some sort had sufficient overhead to
work correctly when close to the 10V threshold?

>That way, the characteristics of the zener knee are important
>over a 5:1 current range instead of an infinite ratio.

ITRW, as long as it's starting to work reasonably by 10.5V it
shouldn't really be an issue.

>I'm far too lazy to do the math on how nonlinear it will be.

I do like the idea of printing my own scale and if this basic one
works, I might go for a bigger / better meter that I can open etc.

>
>There's also an elephant in the room.
>Unless you start with a meter that happens to have scale
>markings ranging from 10 to 15, you're gonna need a new scale
>no matter what you do with the electronics.

Unless it's already marked 0, 1, 2, 3, 4 and 5 'Volts' and you know
just have to add 10 to all of them. I would imaging the needle
pointing up the middle to 12.x would be a sufficient comforter. ;-)

>If you're not gonna
>do that, might as well just mark calibrations on the face with a
>sharpie.

Only if I can get the thing open and it wouldn't be worth damaging it
to try.

>Makes no sense to go to great lengths in one part of the project
>only to create something that still doesn't do what you want.

My only hope is that it works 'better' than a similarly sized and
accuracy meter that is marked 0 to 15 and using it over a tiny
proportion of the range.

As has been suggested, if you aren't happy to try to KISS then you are
then going to be worrying about thermal stability and scale linearity
and I'd like to leave that for the Mk2. ;-)

Cheers, T i m

[T i m]

On Fri, 22 Sep 2017 14:13:13 -0700 (PDT), whit3rd <whi...@gmail.com>
wrote:

>On Thursday, September 21, 2017 at 10:51:33 AM UTC-7, T i m wrote:
>> Hi All,
>>
>> Not a 'repair' question as such but I was thinking of using an old
>> skool 250uA FSD moving coil panel meter to read the 10 to 15V DC range
>> (a lead acid battery charge / discharge indicator), next to a 0-30A
>> Ammeter (inc shunt etc).
>>
>> I was thinking of using either a voltage reference such as
>> LM4040DIZ-10.0 or could I get away with a simple zener (BZX55C?) with
>> associated resistor(s) please?
>
>You can use any zener under 10V to do a 10V zero offset (just divide down the
>input voltage until that 10V input matches the zener knee). So, with an
>arbitrary zener (I'd use a TL431 as a 2.5V reference, they're convenient
>and accurate) and a trimmer acting as voltage divider, it just remains
>to make a range-setting resistor in series with the meter.

Hmm, that's crafty. ;-)

I'm trying to picture how the voltage rise would still be linear but I
think I can see how it would be.

The -ve of the meter is held at some voltage threshold low enough to
ensude a low change in difference between the voltage measured and the
ground. A variable (to start with at least) resistor set as a pd then
sets the upper voltage to give FSD at 15V.

>
>The important thing, is to doodle up a really nifty scale for your meter
>that indicates 10V to 15V, with clear markings, along the pointer's arc,
>maybe with a bunch of subdivision marks.

Yeah, that sounds like a interesting project for a Mk2 meter. ;-)

>took
>a plotter and a bit of custom software to draw the arcs and labels,

I saw the link to some free software for doing such but not had chance
to try it yet.

>then some fiddling with the faceplate of the meter to
>affix it (warning: you need to worry about laser-print ink, paper, glue compatibility).

Ok. ;-)

Cheers, T i m

[mike]

Thereyougo...opamp is a lot more complex than two resistors...or a couple
of AA cells.

[T i m]

On Fri, 22 Sep 2017 10:53:54 -0500, "Dave M" <dgmi...@mediacombb.net>
wrote:

><snip>

>I forgot to mention that for the zener solution, you really need to select
>as low power a zener that you can find, such as 400mW or lower.
>The 1N4104 is a 10V, 250mW zener, with a test current of 1uA, and a max
>current of 25ma, well within your requirements. In stock at Mouser.com.
>

Cheers, Dave, I'll have a look.

As has been mentioned elsewhere, if we used a lower voltage zener (and
as high a tolerance as can be found in the right spec and the right
price *because*) to set the voltage on the -Ve leg of the meter and
then tweak the voltage up via resistors (variable initially) to give a
FSD at 15V then that should be sufficient?

I can't help feeling I've missed something with that solution though,
like the linearity ... ? ;-(

Cheers, T i m

[tabb...@gmail.com]

On Monday, 25 September 2017 04:43:06 UTC+1, T i m wrote:
> On Sun, 24 Sep 2017 17:52:51 -0700, mike <ham...@netzero.net> wrote:

> >On 9/24/2017 6:31 AM, tabbypurr wrote:
> >> On Sunday, 24 September 2017 12:06:45 UTC+1, mike wrote:
> >>
> >>> A series zener needs to draw ZERO current at 10V and still
> >>> be 10V at MUCH higher current ratio. You will have to live
> >>> with some amount of nonlinearity.
> >>
> >> In fact you could re-zero the meter so it reads 10v spot on. Add an extra mark for where it points when unpowered so you can see if it ever decalibrates.
> >>
> >Yes, you can , but it's still nonlinear.
> >It's a matter of how much nonlinearity you can tolerate.
> >It's my assertion that the measured quantity is so crude that you don't
> >need to worry much about anything. Just learn what it reads and calibrate
> >your expectations to match.
>
> (I've been a bit preoccupied elsewhere but still following the thread
> with interest and will reply individually when I get the chance)
>
> That was part I had already considered in that:
> The meter is fairly small. The meter is glued shut. The meter is only
> marked in whole volts.

If it's not an accurate meter just add a zener, job done.


NT

[tabb...@gmail.com]

On Monday, 25 September 2017 10:01:42 UTC+1, mike wrote:

I'm pretty sure we all know that, and what the options are, and what the pros & cons are. And that adding AA cells rather than 2 resistors on the input is not an option with any upside.


NT

[T i m]

On Mon, 25 Sep 2017 03:41:49 -0700 (PDT), tabb...@gmail.com wrote:

<snip>

>> That was part I had already considered in that:
>> The meter is fairly small. The meter is glued shut. The meter is only
>> marked in whole volts.
>
>If it's not an accurate meter just add a zener, job done.

Well, I'm not sure if it's electro mechanically any more or less
'accurate' than any similar style meters and being smaller with a
shorter / lighter needle it's possible it could be better than some
... but regarding the potential display markings and readability, you
may well be right. ;-)

Cheers, T i m

[tabb...@gmail.com]

On Monday, 25 September 2017 12:37:05 UTC+1, T i m wrote:

If it's only marked in whole volts, reading it with accuracy won't happen, so it won't be especially accurate.


NT

[pf...@aol.com]

On Monday, September 25, 2017 at 10:13:48 AM UTC-4, tabb...@gmail.com wrote:

> If it's only marked in whole volts, reading it with accuracy won't happen, so it won't be especially accurate.

How so? I suggest that you are mistaking "accuracy" for "precision". They are not even a little bit the same thing nor follow necessarily one from another.

A gauge or meter that reads (is marked) in whole numbers but is dead-on is far more accurate than a gauge or meter marked in thousandths, but is always two units off. Although the latter meter is extremely precise.

I have a metered Iso-Variac that reads in volts measured in tens, and amps measured in tenths on a 0 -1 or 0 - 4 scale. But they are 4" meters so that even very slight needle movements are visible. The human brain is pretty good at interpolation.

[Dave M]

I gave the wrong zener current for the 1N4104 . I just looked at Mouser's
basic info for the diode, but neglected to follow up and look at the actual
datasheet. According to the datasheet, the zener test current is actually
250uA, whereas Mouser states it as 1uA. Quite a difference, but after
looking at the datasheets for some other 10V/200mW zeners, (datasheets that
had actual graphs for Iz vs. Vz) I see that they actually start conducting
well below 1uA, so I feel confident that the 1N4104 will do the job for you.
All the other zeners that I found are SMD devices, so I doubt that you'll be
interested in any of those.
With that Zener in the circuit, I doubt that you'll have any noticible
nonlinearity; certainly not with the meter scale only having a 1-volt
resolution. Forget about the accuracy on the low end as well (except at the
volt marks), because you can't resolve to a basic meter accuracy
specification of +-3% of FSD.
Dave M

[T i m]

On Mon, 25 Sep 2017 07:53:49 -0700 (PDT), "pf...@aol.com"
<pf...@aol.com> wrote:

>On Monday, September 25, 2017 at 10:13:48 AM UTC-4, tabb...@gmail.com wrote:
>
>> If it's only marked in whole volts, reading it with accuracy won't happen, so it won't be especially accurate.
>
>How so? I suggest that you are mistaking "accuracy" for "precision". They are not even a little bit the same thing nor follow necessarily one from another.

+1

>
>A gauge or meter that reads (is marked) in whole numbers but is dead-on is far more accurate than a gauge or meter marked in thousandths, but is always two units off.

Yup, that's the thing ... and why many battery meters are calibrated
with red, amber and green sections. Now whilst that would work for me
re charge (voltage) and loosely therefore charge status, it wouldn't
be useful to pin down the discharge / charge status (unlike say a CCA
tester) as the terminal volts are a function of load.

> Although the latter meter is extremely precise.

Yup and I have some of those. Highly repeatable but always 'out' by
some amount. ;-(

>
>I have a metered Iso-Variac that reads in volts measured in tens, and amps measured in tenths on a 0 -1 or 0 - 4 scale. But they are 4" meters so that even very slight needle movements are visible. The human brain is pretty good at interpolation.

And it was the latter thought that I was using when it came to this
meter. ;-)

Take something that offers a very 'human' way of displaying something
(like an analogue car fuel or specifically, temperature gauge that
starts at say 60 DegC) and retain the merits but improve the general
meaning / readability.

Mileage / trip counters are more suited to being 'digital' etc. ;-)

Cheers, T i m

[tabb...@gmail.com]

On Monday, 25 September 2017 15:53:59 UTC+1, pf...@aol.com wrote:

With very coarse markings you won't get accurate readings. End of relevant material. I have no motivation to argue over the simple and immaterial.


NT

[Foxs Mercantile]

On 9/25/2017 3:35 PM, Dave M wrote:
> I see that they actually start conducting well below 1uA, so I feel
> confident that the 1N4104 will do the job for you.
> All the other zeners that I found are SMD devices, so I doubt that
> you'll be interested in any of those.

Previously I wrote:

The simplest is a 20K resistor in series with a 10v zener diode.
Use a low current Zener so the knee works right at low values.
Something like this:
<http://www.mouser.com/ds/2/395/BZX55C2V0%20SERIES_D1610-1099630.pdf>
Note, this is a DO-35 package.


--
Jeff-1.0
wa6fwi
http://www.foxsmercantile.com

[whit3rd]

On Sunday, September 24, 2017 at 8:43:06 PM UTC-7, T i m wrote:
> On Sun, 24 Sep 2017 17:52:51 -0700, mike <ham...@netzero.net> wrote:

> >I'm far too lazy to do the math on how nonlinear it will be.
>
> I do like the idea of printing my own scale and if this basic one
> works, I might go for a bigger / better meter that I can open etc.

Yep, that's a good idea.

And, there's NO NONLINEARITY unless you apply bias less than the zener
voltage breakdown, because of Norton's theorem: any bunch of
resistors and voltage sources (the zener is a voltage source) is
identically a current source with a shunt resistance. Offset by
a constant, you can get; nonlinearity, you can't. The meter magnetic field
is the only source of nonlinearity (and meter manufacturers do pretty well
on keeping that uniform).

[tabb...@gmail.com]

zener v/i curves vary, some are terrible near zero i, some straighter.


NT

[T i m]

On Mon, 25 Sep 2017 20:08:01 -0500, Foxs Mercantile <jda...@att.net>
wrote:

So, if the / any issue to the zener solution is the closeness of any
knee to the bottom end of the required voltage range, would the using
the / a lower reference and a potential devider to set the fsd on 15
volts be a better idea still (or have I missed something again)? ;-(

Cheers, T i m

[T i m]

On Mon, 25 Sep 2017 15:35:28 -0500, "Dave M" <dgmi...@mediacombb.net>

wrote:

>T i m wrote:
>> On Fri, 22 Sep 2017 10:53:54 -0500, "Dave M" <dgmi...@mediacombb.net>
>> wrote:
>>
>>> <snip>
>>
>>> I forgot to mention that for the zener solution, you really need to
>>> select as low power a zener that you can find, such as 400mW or
>>> lower.
>>> The 1N4104 is a 10V, 250mW zener, with a test current of 1uA, and a
>>> max current of 25ma, well within your requirements. In stock at
>>> Mouser.com.
>>>
>> Cheers, Dave, I'll have a look.
>>
>> As has been mentioned elsewhere, if we used a lower voltage zener (and
>> as high a tolerance as can be found in the right spec and the right
>> price *because*) to set the voltage on the -Ve leg of the meter and
>> then tweak the voltage up via resistors (variable initially) to give a
>> FSD at 15V then that should be sufficient?
>>
>> I can't help feeling I've missed something with that solution though,
>> like the linearity ... ? ;-(
>>
>

>I gave the wrong zener current for the 1N4104 . I just looked at Mouser's
>basic info for the diode, but neglected to follow up and look at the actual
>datasheet. According to the datasheet, the zener test current is actually
>250uA, whereas Mouser states it as 1uA. Quite a difference, but after
>looking at the datasheets for some other 10V/200mW zeners, (datasheets that
>had actual graphs for Iz vs. Vz) I see that they actually start conducting
>well below 1uA, so I feel confident that the 1N4104 will do the job for you.

Thanks for the follow up Dave.

>All the other zeners that I found are SMD devices, so I doubt that you'll be
>interested in any of those.

Well, I have worked with SMD's but a wired device and could use on at
a pinch (soldered to the side of the meter contact etc) bit a wired
device is probably easier.

>With that Zener in the circuit, I doubt that you'll have any noticible
>nonlinearity; certainly not with the meter scale only having a 1-volt
>resolution. Forget about the accuracy on the low end as well (except at the
>volt marks), because you can't resolve to a basic meter accuracy
>specification of +-3% of FSD.

Understood. I was thinking more on the idea of when using a lower
voltage zener to be well clear of the lower voltage levels being
considered and then the potential divider to set the FSD when the
input is 15V?

And also, wouldn't the impact of the battery voltage changing from say
10.5 to 14.4V be lessened (re the series current limiting / biasing
resistor) if the voltage difference between the reference and the
supply is greater?

That seemed like the best of all worlds (for the KISS solution)?

Cheers, T i m

[T i m]

Doesn't that depend what you are looking for? Say I determine my LVD
should be at 11.000 volts then I would appreciate the meter being as
accurate as possible (at that point at least)?

Cheers, T i m

[tabb...@gmail.com]

You want a zener with a nice straight line at the low current end. One linked to from here has that.


NT

[T i m]

On Tue, 26 Sep 2017 06:46:01 -0700 (PDT), tabb...@gmail.com wrote:

<snip>

>> Understood. I was thinking more on the idea of when using a lower
>> voltage zener to be well clear of the lower voltage levels being
>> considered and then the potential divider to set the FSD when the
>> input is 15V?
>>
>> And also, wouldn't the impact of the battery voltage changing from say
>> 10.5 to 14.4V be lessened (re the series current limiting / biasing
>> resistor) if the voltage difference between the reference and the
>> supply is greater?
>>
>> That seemed like the best of all worlds (for the KISS solution)?

>

>You want a zener with a nice straight line at the low current end. One linked to from here has that.
>

Yeah, ok ... but what about the question re using a lower value zener
to get it away from the 'knee', especially as it's no more difficult
to do?

Cheers, T i m

[Foxs Mercantile]

Here it is again.

The simplest is a 20K resistor in series with a 10v zener diode.
Use a low current Zener so the knee works right at low values.
Something like this:
<http://www.mouser.com/ds/2/395/BZX55C2V0%20SERIES_D1610-1099630.pdf>
Note, this is a DO-35 package.

[mike]

Since we're just trying to be clever,
Try this:

https://i.imgur.com/irLisN7.jpg

The transistor and zener make a temperature compensated ~7V
reference (must use 6.2V zener with standard tempco).
Two pots set the min and max voltages. Rk allows you to add
current to the zener to get it past the most nonlinear portion
of it's V-I curve. This circuit has the ability to calibrate out
almost any component variation. It's relatively temperature
independent. You can use any meter that will make full scale at
less than about 7V. Downside is that the pots are vibration sensitive
and subject to moisture/corrosion.

Are we having fun yet?

[tabb...@gmail.com]

On Tuesday, 26 September 2017 17:26:46 UTC+1, T i m wrote:

The zeners with the straightest knee are at 9-10v. Go look at the linked data sheets.


NT

[T i m]

On Tue, 26 Sep 2017 14:13:20 -0700, mike <ham...@netzero.net> wrote:

<snip>

>> Yeah, ok ... but what about the question re using a lower value zener
>> to get it away from the 'knee', especially as it's no more difficult
>> to do?
>>
>>

>Since we're just trying to be clever,

I'm not mike, I'm just trying to refine the design of any solution to
be as good as possible with the fewest possible components. ;-)

>Try this:
>
>https://i.imgur.com/irLisN7.jpg
>
>The transistor and zener make a temperature compensated ~7V
>reference (must use 6.2V zener with standard tempco).

Ok.

>Two pots set the min and max voltages.

Noted.

>Rk allows you to add
>current to the zener to get it past the most nonlinear portion
>of it's V-I curve.

Understood.

>This circuit has the ability to calibrate out
>almost any component variation.

That sounds like the sort of thing I was hoping to hear. ;-)

>It's relatively temperature
>independent.

As long as it will cope with the -10 to +10 DegC range it should be
fine here in the UK. ;-)

>You can use any meter that will make full scale at
>less than about 7V.

And that's most small panel meters I'm sure.

>Downside is that the pots are vibration sensitive
>and subject to moisture/corrosion.

Sure, but in my case they won't be subject to any real vibration and
will be in a sealed enclosure. Also I would typically use a quality
multiturn pot of the smallest value possible to give me sufficient
'trimming' around the end points and quality resistors to do the main
bridge / work.

>
>Are we having fun yet?

Well, it's certainly bringing some interesting and varied solutions
out of the woodwork. ;-)

It might also help me use up my stock of BC107's. ;-)

Cheers, T i m

[T i m]

On Tue, 26 Sep 2017 15:29:19 -0700 (PDT), tabb...@gmail.com wrote:

<snip>

>> Yeah, ok ... but what about the question re using a lower value zener
>> to get it away from the 'knee', especially as it's no more difficult
>> to do?
>>
>

>The zeners with the straightest knee are at 9-10v.

Ok, so again that helps because I'd go for a 9V zener to get AFA I can
from that 10V threshold and still retaining the sharpest knee. ;-)

>Go look at the linked data sheets.

Too much information (variables) for me in most cases mate. ;-)

Ah, you can use the 9V zener but it only comes in SM and you have to
buy 1000. The 10V are easier to get but need a higher current ... etc
etc. I don't play chess or poker or any game of chance for that matter
(preferring FPS when I can shoot my fellow players in the face). ;-)

The whole point to me of asking those who know is they are the ones
who should all agree on a single 'best' solution. ;-)

More than one way to skin a cat? (But for that as long as it does get
skinned is good enough for me [1]).


Cheers, T i m

[1] I couldn't hurt one but I don't see the point in them ... apart
from a source of whiskers for crystal sets and guts for violins. ;-)

[Foxs Mercantile]

On 9/26/2017 7:49 PM, T i m wrote:
> Too much information (variables) for me in most cases mate.;-)

You keep missing it.

[tabb...@gmail.com]

On Wednesday, 27 September 2017 01:49:50 UTC+1, T i m wrote:

Just use a 10v zener. Job done. There's little point in the rest when the meter is so coarsely marked. You can adjust for 10v mechanically. You should find one in your scraps box.

Catgut is made from sheep, goats etc.


NT

[mike]

Use a small value pot with a fixed resistor on each end.
Easier to set, more stable.