Resistor ratios

[George Herold]

On Wednesday, March 2, 2016 at 2:51:55 PM UTC-5, John Larkin wrote:

On Wed, 02 Mar 2016 12:23:52 -0600, Tim Wescott
seemywebsite@myfooter.really> wrote:

On Wed, 02 Mar 2016 05:54:56 -0800, John Larkin wrote:

On Tue, 01 Mar 2016 10:41:18 -0600, Tim Wescott <tim@seemywebsite.com
wrote:

On Mon, 29 Feb 2016 20:36:11 -0800, John Larkin wrote:

On Mon, 29 Feb 2016 21:40:22 -0600, Tim Wescott <tim@seemywebsite.com
wrote:

On Mon, 29 Feb 2016 17:31:09 -0800, John Larkin wrote:

On Mon, 29 Feb 2016 16:49:53 -0800 (PST), George Herold
gherold@teachspin.com> wrote:

On Monday, February 29, 2016 at 2:12:40 PM UTC-5, Phil Hobbs wrote:
On 02/29/2016 11:52 AM, George Herold wrote:
Hi all, I'm making this current output in fixed ratio's...
1,2,4...
etc.
I'm using 0.1% RG series from susumu.
There are lots of 1:2 ratio's,
But the only 1:2:4 that I find (in stock at DK)
is 75/150/300. are there any others?

(There is 140/280/560, but 280 is a non-stock item at DK.)

75/150/300 also has 1.2K so a 1:2:4..16 ratio, which look nice.

George H.

Well, 249/499/1000/2000 is close, and 255/510/1020 is nominally
exact.
(Susumu sells the E24 values in their precision series as well,
iirc.)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs Principal Consultant ElectroOptical
Innovations LLC Optics, Electro-optics, Photonics, Analog
Electronics

160 North State Road #203 Briarcliff Manor NY 10510

hobbs at electrooptical dot net http://electrooptical.net

Tim, Phil, thanks. I had a total brain fart today.
It's an opamp -> fet -> load/resistor_feedback current source.
I wanted to switch R (with rotary switch) and I thought I could do
it in shunt/ parallel, until I tried to lay it out.. oops. (I hope
the odor did not escape my room.)

Max current is in the 200mA range. (limited by switch...)
Now I'm thinking about ditching the rotary switch and using a
latching relays.. ~$2-3 each (in 100)
Seems like I could do a lot with 3-4.
I'd still have to have inputs.. push buttons I guess..
and led indicators. A front panel pcb...
(rotary switches are on the shelf.)

Sorry for thinking out loud.

George H .

PS,
Phil, Re: 249,499, 1k, 2k, 4.02k I wish we could "bend the log a
bit" and have 250,500,1k,2k,4k..

GH




Can you do the current control on the input side, rather than
switching 200 mA? Use an analog mux or a low-current rotary switch.

Does anybody still make R-2R ladder networks? I guess you could make
your own.

Use a DAC?

DigiKey has a bunch of DAC0808s in stock. It looks like it's not the
least bit dead.

There are tons of DACs around, so George could do something to map his
rotary switch into DAC codes. Actually, it's simple 1:1 if he wants
2:1 current steps.

A chopamp would have essentially zero offset error.

You and I are thinking alike here.

There is something steampunk and nostalgic about rotary switches. But
the switch and knob can get expensive these days, not to mention wiring.

I know. It's lowering, but every time I end up doing a full-system
design the case, knobs and buttons end up costing more than the circuitry.

Well, buy more expensive parts!
Grin.. in my world opamps are almost free.

We've found that designing our own box, and having a shop
punch/bend/anodize the sheet metal, is cheaper than buying a
commercial box, much less machining it.
Yeah you need to get up above the ~100 level.

We make these "bath tubs" milled out of Al bar stock,
Sort of 1.5" X 1.5" x 3". a bit bigger than a small blue
pomona box. $2-3 each.. or something like that.
(I guess there is a powder coating step too.)

George H.

--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[John Larkin]

On Thu, 3 Mar 2016 06:56:20 -0800 (PST), George Herold
<gherold@teachspin.com> wrote:

On Wednesday, March 2, 2016 at 2:51:55 PM UTC-5, John Larkin wrote:
On Wed, 02 Mar 2016 12:23:52 -0600, Tim Wescott
seemywebsite@myfooter.really> wrote:

On Wed, 02 Mar 2016 05:54:56 -0800, John Larkin wrote:

On Tue, 01 Mar 2016 10:41:18 -0600, Tim Wescott <tim@seemywebsite.com
wrote:

On Mon, 29 Feb 2016 20:36:11 -0800, John Larkin wrote:

On Mon, 29 Feb 2016 21:40:22 -0600, Tim Wescott <tim@seemywebsite.com
wrote:

On Mon, 29 Feb 2016 17:31:09 -0800, John Larkin wrote:

On Mon, 29 Feb 2016 16:49:53 -0800 (PST), George Herold
gherold@teachspin.com> wrote:

On Monday, February 29, 2016 at 2:12:40 PM UTC-5, Phil Hobbs wrote:
On 02/29/2016 11:52 AM, George Herold wrote:
Hi all, I'm making this current output in fixed ratio's...
1,2,4...
etc.
I'm using 0.1% RG series from susumu.
There are lots of 1:2 ratio's,
But the only 1:2:4 that I find (in stock at DK)
is 75/150/300. are there any others?

(There is 140/280/560, but 280 is a non-stock item at DK.)

75/150/300 also has 1.2K so a 1:2:4..16 ratio, which look nice.

George H.

Well, 249/499/1000/2000 is close, and 255/510/1020 is nominally
exact.
(Susumu sells the E24 values in their precision series as well,
iirc.)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs Principal Consultant ElectroOptical
Innovations LLC Optics, Electro-optics, Photonics, Analog
Electronics

160 North State Road #203 Briarcliff Manor NY 10510

hobbs at electrooptical dot net http://electrooptical.net

Tim, Phil, thanks. I had a total brain fart today.
It's an opamp -> fet -> load/resistor_feedback current source.
I wanted to switch R (with rotary switch) and I thought I could do
it in shunt/ parallel, until I tried to lay it out.. oops. (I hope
the odor did not escape my room.)

Max current is in the 200mA range. (limited by switch...)
Now I'm thinking about ditching the rotary switch and using a
latching relays.. ~$2-3 each (in 100)
Seems like I could do a lot with 3-4.
I'd still have to have inputs.. push buttons I guess..
and led indicators. A front panel pcb...
(rotary switches are on the shelf.)

Sorry for thinking out loud.

George H .

PS,
Phil, Re: 249,499, 1k, 2k, 4.02k I wish we could "bend the log a
bit" and have 250,500,1k,2k,4k..

GH




Can you do the current control on the input side, rather than
switching 200 mA? Use an analog mux or a low-current rotary switch.

Does anybody still make R-2R ladder networks? I guess you could make
your own.

Use a DAC?

DigiKey has a bunch of DAC0808s in stock. It looks like it's not the
least bit dead.

There are tons of DACs around, so George could do something to map his
rotary switch into DAC codes. Actually, it's simple 1:1 if he wants
2:1 current steps.

A chopamp would have essentially zero offset error.

You and I are thinking alike here.

There is something steampunk and nostalgic about rotary switches. But
the switch and knob can get expensive these days, not to mention wiring.

I know. It's lowering, but every time I end up doing a full-system
design the case, knobs and buttons end up costing more than the circuitry.

Well, buy more expensive parts!
Grin.. in my world opamps are almost free.

We've found that designing our own box, and having a shop
punch/bend/anodize the sheet metal, is cheaper than buying a
commercial box, much less machining it.
Yeah you need to get up above the ~100 level.

We make these "bath tubs" milled out of Al bar stock,
Sort of 1.5" X 1.5" x 3". a bit bigger than a small blue
pomona box. $2-3 each.. or something like that.
(I guess there is a powder coating step too.)

George H.

This box is nice, looks pretty good and is easy to work on.

http://www.highlandtechnology.com/DSS/P350DS.shtml

http://www.highlandtechnology.com/DSS/P470DS.shtml

It winds up costing us about $100 in modest quantities.


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com