0.01% resistors

[John Devereux]

As far as I can see that doesn't work if the output is shorted.

No, although the output will indicate some high full scale value I
think.

You did say it was around 16V...

If it really has to work 0-16V, and you also need 1mV precision, then
there is not getting away from the big bucks resistor.

Or does the 16V supply collapse too when it is shorted? Like you are
measuring current using a high side resistor on the output side of a
current limited regulator?

Battery charger?

You could look at the "flying capacitor" circuit as found in that linear
app note.

 

[Spehro Pefhany]

Only if you know enough about what you are posting to avoid posting duff information. Some Wikipedia articles, like some application notes, are not to be relied on.

At least you are free to fix Wikipedia.


Best regards,
Spehro Pefhany

 

[Bill Sloman]

As far as I can see that doesn't work if the output is shorted.

So your Howland circuit runs off completely independent power rails - apartfrom sharing a common ground.

Then you should take a serious look at an instrumentation amplifier, drivenfrom the same independent rails.

 

[Spehro Pefhany]

Good point. I'll check it later tonight.


Best regards,
Spehro Pefhany

This looks better;

# Circuit Ratio LM317 LT1964 LT1761 TL431
0 -- 1.000 1.25 -1.22 1.22 2.50
13 "(a|b) + (c+d)", 0.800 1.56 -1.53 1.53 3.12
19 "(b | (a+c)) + d FB a-c" 0.800 1.56 -1.53 1.53 3.12
4 "a+ (b+c+d)", 0.750 1.67 -1.63 1.63 3.33
2 "a + (b+c)", 0.667 1.88 -1.83 1.83 3.74
7 "(a|b) + c", 0.667 1.88 -1.83 1.83 3.74
9 "a + (b + (c|d))", 0.600 2.08 -2.03 2.03 4.16
11 "a + ((b|c)+d)", 0.600 2.08 -2.03 2.03 4.16
17 "(b|(a+c))+d", 0.600 2.08 -2.03 2.03 4.16
1 "a+b", 0.500 2.50 -2.44 2.44 4.99
5 "(a+b) + (c+d)", 0.500 2.50 -2.44 2.44 4.99
15 "((a|b) + (c|d))", 0.500 2.50 -2.44 2.44 4.99
12 "(a+(b|c)) + d", 0.400 3.13 -3.05 3.05 6.24
14 "((a|b)+c) + d", 0.400 3.13 -3.05 3.05 6.24
3 "(a+b) + c", 0.333 3.75 -3.66 3.66 7.49
8 "a + (b|c)", 0.333 3.75 -3.66 3.66 7.49
6 "(a+b+c) + (d)", 0.250 5.00 -4.88 4.88 9.98
10 "(a+b) + (c|d)", 0.200 6.25 -6.10 6.10 12.48
16 "(a) + (b|(c+d))", 0.200 6.25 -6.10 6.10 12.48
18 "a + (b | (c+d)) FB c-d" 0.200 6.25 -6.10 6.10 12.48

Best regards,
Spehro Pefhany

 

[[email protected]]

If it really has to work 0-16V, and you also need 1mV precision, then
there is not getting away from the big bucks resistor.

Or does the 16V supply collapse too when it is shorted? Like you are
measuring current using a high side resistor on the output side of a
current limited regulator?

Battery charger?

You could look at the "flying capacitor" circuit as found in that linear
app note.

Another option is an auxiliary over-the-top supply. Then you can
use a low-side sensing technique, and reflect the output off the
ground rail.

A little messy parts-wise, but simple, solid, and accurate.

 

[John Devereux]

Cool. What are those FB minus things?

My original was in context of making an opamp booster stage.

Resistor network on the opamp output to ground. Feedback to opamp - from
some point in the network. The FB indicated where the feedback was taken
from, where this could not be indicated by bracketing.

 

[Spehro Pefhany]

Cool. What are those FB minus things?



From John D's nomenclature- eg.on the bottom one you take the feedback
(adj pin) from the junction of the two series resistors c-d.

Other combinations give the same ratio when all the resistors are equal,
but maybe the series resistance or divider output resistance is more
suitable in one case than another.

Best regards,
Spehro Pefhany

 

[Fred Abse]

Anyone know what are these "Taitanium-Nitrate" arrays? I think they mean
tantalum nitride

It says "TaN" on the schematic. I think you're right.

 

[josephkk]

This looks better;

# Circuit Ratio LM317 LT1964 LT1761 TL431 0 --
1.000 1.25 -1.22 1.22 2.50 13 "(a|b) + (c+d)", 0.800
1.56 -1.53 1.53 3.12 19 "(b | (a+c)) + d FB a- c" 0.800 1.56
-1.53 1.53 3.12 4 "a+ (b+c+d)", 0.750 1.67 -1.63
1.63 3.33 2 "a + (b+c)", 0.667 1.88 -1.83 1.83
3.74 7 "(a|b) + c", 0.667 1.88 -1.83 1.83 3.74
9 "a + (b + (c|d))", 0.600 2.08 -2.03 2.03 4.16 11 "a
+ ((b|c)+d)", 0.600 2.08 -2.03 2.03 4.16 17
"(b|(a+c))+d", 0.600 2.08 -2.03 2.03 4.16 1 "a+b",
0.500 2.50 -2.44 2.44 4.99 5 "(a+b) + (c+d)",
0.500 2.50 -2.44 2.44 4.99 15 "((a|b) + (c| d))", 0.500
2.50 -2.44 2.44 4.99 12 "(a+(b|c)) + d", 0.400 3.13
-3.05 3.05 6.24 14 "((a|b)+c) + d", 0.400 3.13 -3.05
3.05 6.24 3 "(a+b) + c", 0.333 3.75 -3.66 3.66
7.49 8 "a + (b|c)", 0.333

3.75 -3.66 3.66 7.49 6

"(a+b+c) + (d)", 0.250 5.00 -4.88 4.88 9.98 10 "(a+b)
+ (c|d)", 0.200 6.25 -6.10 6.10 12.48 16 "(a) +
(b|(c+d))", 0.200 6.25 -6.10 6.10 12.48 18 "a + (b |
(c+d)) FB c-d" 0.200 6.25 -6.10 6.10 12.48

Best regards,
Spehro Pefhany

I don't know why, but the table is totally mangled in pan.

?-(

 

[John Devereux]

3.75 -3.66 3.66 7.49 6

I don't know why, but the table is totally mangled in pan.

In his OP the column alignment was done with a mixture of spaces and
tabs (which are viewer-dependent). Then in your reply it is line wrapped
too.

If you can still see the OP you could try turning off line wrapping and
fiddling with the tab width.

 

[Glenn]

3.75 -3.66 3.66 7.49 6

I don't know why, but the table is totally mangled in pan.

?-(

It is because the tab-char to the number og space-chars is not unique.

Some editor, mail clients, word process. might use e.g. 4space:1tab or
4space:1tab.

It is better that the sender only use space character - combined with a
fixed char width font.

/Glenn

 

[Glenn]

3.75 -3.66 3.66 7.49 6

I don't know why, but the table is totally mangled in pan.

?-(

It is because the tab-char to the number of space-chars is not unique.

Some editor, mail clients, word process. might use e.g. 4space:1tab or
8space:1tab.

It is better that the sender only use space character - combined with a
fixed char width font.

/Glenn

 

[Fred Abse]

I don't know why, but the table is totally mangled in pan.

View - Wrap article body - make sure it's OFF.

Displays fine in pan here...

 

[josephkk]

View - Wrap article body - make sure it's OFF.

Displays fine in pan here...

Thanks Fred, that untrashed it.

?-)

 

[Jasen Betts]

Someone should work out all the voltages possible from an LM317 and a quad
r-pack.

assuming 4 equal R's you need 2 finite resistors to set the voltage so

3Rs can make: C= {1/3,1/2,2/3,1,1.5,2,3}
2Rs can make: B= {1/2,1,2}
1R can make: A= {1}

the resuting ratios are

AxC U BxB U CxA


AxC = 1/3:1 1/2:1 2/3:1 1:1 1.5:1 2:1 3:1

= 1:3 1:2 2:3 1:1 3:2 2:1 3:1

BxB = 1/2:1/2 1:1/2 2:1/2
1/2:1 1:1 2:1
1/2:2 1:2 2:2

= 1:4 1:2 1:1 2:1 4:1

CxA = 1:1/3 1:1/2 1:2/3 1:1 1:1.5 1:2 1:3

3:1 2:1 3:2 1:1 2:3 1:2 1:3

(these ratios are the same as AxC - just reordered)


so the whole set is

1:4 1:3 1:2 2:3 1:1 3:2 2:1 3:1 4:1

and the no resistor state 0:inf

---[317]----+--
| |
|--[y]-'
|
[x]
|
---+--- 0v

so for boosting a 1.25V regulator you get

x:y V

0:inf 1.250
1:4 1.563
1:3 1.667
1:2 1.875
2:3 2.083
1:1 2.500
3:2 3.125
2:1 3.75
3:1 5.00
4:1 6.25


or you could do this,

---[317]-+-- v
| |
| R
| |
+-R-+ <-- extra node
| |
R R
| |
---+---+---

For umm, looks like 1:4 again
---[317]-+-- V
| |
| R 3i
| i |
+-R-+
| |
i R R 2i
| |
---+---+---

Yep.

or this,

---[317]-+-- V
| |
| i |
+-R-+
| |
| R 2i
| i |
+-R-+ <-- extra
|
R 3i
|
---+---+---

4:1 again. 6.25V


hmm, that means there's atleast two ways to do each ratio

 

[Don Lancaster]

Just leafing tghrough a Venkel catalog, they have surfmount 0.01%, 5
PPM resistors.

Without a four point Kelvin connection, there is no way in hell you can
get 0.01 percent in-circuit accuracy.

--
Many thanks,

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

Please visit my GURU's LAIR web site at http://www.tinaja.com