4

I'm looking at building my own decade resistance box, but the typical rotary switches are... quite expensive. They seem to be the most expensive part of the entire project. Whether I choose a real rotary switch or a DIP switch or even a thumbwheel switch, the standard decade resistance box uses an SP10T switch that is either expensive or limited in selection.

I've seen a lot more switches that use a BCD output code, or a BCD complement code (example: page six of this datasheet, specifically switch functions 1 and 2. But I'm having a tough time wrapping my head around the connections. In an SP10T switch, you connect the 0 terminal to the "input", and the common terminal is the "output" (or vice versa), as seen here: credit to eevblog 212 (source: EEVblog #212)

But a BCD switch makes multiple connections depending on the desired code. For example, if I set the switch to 3, then the common terminal will be shorted to positions 1 and 2. In a BCD-complement switch, if I set the switch to 3, then the common terminal is shorted to positions 4 and 8. Either way, I can't come up with an arrangement of resistors that would yield a resistance of 3ohms (or 30, or 300, etc). This is the switch function diagram for the switches I have in mind, which might be a clearer explanation of the behavior I'm describing: c&k datasheet

Sorry if this question is kind of confusing, I'm having a hard time explaining it myself! To summarize, does anyone have a connection diagram for a decade resistance box using BCD switches?

Community
  • 1
tummychow
  • 651
  • 2
  • 9
  • 18

4 Answers4

1

For a BCD switch, connect resitors wil values of 1, 2, 4, and 8 ohms (or decade multiples thereof) in series, with a switch in parallel with each resistor - the switches short the unwanted resistor.

Like so:

schematic

simulate this circuit – Schematic created using CircuitLab

Beware! this will let you set resistances up to 1500 ohms for this decade.

Peter Bennett
  • 59,212
  • 1
  • 49
  • 132
  • This is what I was originally thinking, but I took a look at the connections diagram here: http://puu.sh/4tluI/f7baad4658.png (C&K datasheet) and I'm not sure it'll work as intended. If I use a BCD switch in that configuration, then right off the bat it fails, because if I set it to 0 (all switches open), I get 1500 ohms. If I use a BCD-complement switch, then it'll work in the zero position, because all the switches will be closed. (Continued in next comment...) – tummychow Sep 17 '13 at 03:23
  • However, I was under the impression that the switch looks like this, not like what you've drawn. If that's the case (again, I might be misunderstanding), then when you set the switch to 1, you still have a short from common to 2/4/8, so the result is a short. (continued in next comment...) – tummychow Sep 17 '13 at 03:33
  • In other words, I don't think the four switches inside the BCD switch are connected in a series daisy-chain, but rather in parallel. Am I just misunderstanding the connection diagram? – tummychow Sep 17 '13 at 03:35
  • @tummychow: For the arrangement I drew, you would need four separate switches (or a 4-pole DIP switch), not a BCD switch. – Peter Bennett Sep 17 '13 at 03:42
  • Alright, but I'm specifically looking at the types of DIP/rotary switch that have a set position from 1-10, which is output as a BCD code. 4 separate switches is not really applicable to this situation. I'll add that detail to my question in case that wasn't clear. – tummychow Sep 17 '13 at 03:51
  • Using steps 1, 2, 3, 4 seems to be quite common: http://gmetechnology.com/index.php/products/test-and-measurement-equipment/rd-300-resistance-decade-box – starblue Sep 17 '13 at 07:35
0

Original quote - "For a BCD switch, connect resitors wil values of 1, 2, 4, and 8 ohms (or decade multiples thereof) in series, with a switch in parallel with each resistor - the switches short the unwanted resistor."

CAUTION: Most BCD thumbwheel switches do not work this way. They connect from common to 1 - 4 terminals based on the code. That is, if switch is set at 1, then common is connected to switch terminal 1. If 2, then common is connected to terminal 2. The challenge is for numbers such as 3, where terminal 1 and 2 are connected to common, that is 1 and 2 are in parallel. There is no way to create a decade resistance box with standard BCD rotary switches! They are great for capacitance decades though, as capacitors in parallel add up. So 1, 2, 4, 8 xF capacitors will create a 1-9 xF plus decade (x - any unit, pF, nF, uF).

user46596
  • 1
0

Instead of using BCD Thumb / Push wheel switch one can use DecimalThumb / Push wheel switch with extended PCB at back end where 10 equal resistors can be connected in series. Please watch following video on You:https://www.youtube.com/watch?v=W5kov-MQceY

Umesh Dodia
  • 1
0

This might help

I know, they're in Australia, but it's not like Jaycar make them; one of the big international vendors like Digikey, Mouser, RS, Farnell, etc. will sell similar, if not identical, switches.

markt
  • 4,956
  • 2
  • 15
  • 13
  • Looks to me like a SP12T rotary switch? Wish they provided a datasheet. The price is right, but the only comparable digikey part I can find is this. I admit, that might be able to do the job of making a decade resistance box, but I can source BCD switches for cheaper (after factoring in the cost of the shaft for the comparable SP10T part). – tummychow Sep 17 '13 at 08:13
  • It's configurable, SP2T to SP12T, using a pin that sets the rotation limit. Search digikey for "SP12T". This one is pretty close to the one from Jaycar: http://www.digikey.com.au/product-detail/en/KC52A30.001NPS/EG1951-ND/101787 – markt Sep 17 '13 at 09:04
  • Yes, but that part is quite expensive, as are many 10-throw rotary switches. Again, I want to use a BCD or BCD complement switch, not an SPxT switch (x being some number), if at all possible. For whatever reason, the BCD ones are cheaper and there's more selection. I already know how to wire up a decade resistance box with an SP10T switch; it's the BCD/BCD-complement ones that I'm concerned with. – tummychow Sep 18 '13 at 07:20