Mar 10, 2015, 09:14 AM Hamburger Joined Jan 2010 918 Posts Question how do I reduce current for a given potentiometer when reading voltage? Hi, the topic was maybe over simplistic and not very clear, so here goes Given: I have a potentiometer of 25 Ohm , battery voltage of 5V. Target is to measure the position of potentiometer by observing some voltage in the possible range of [ 0 ; 5] Volts. The current (=heat) through the potentiometer should be low. Initial idea was to use the potentiometer as a variable voltage divider. Drawback is the relatively high current of 5V/25 Ohm = 200mA. (the potentiometer is below the 1W type, so not good) Now I am looking for a way to achieve same task but with lower current. Observing the output voltage is done with an arduino analogRead() if that matters. Using more resistors in series with the potentiometer would kinda work, but would reduce the usable range on the output side and in effect would reduce resolution. So not good. It is probably something simple like controlling the base current of a transistor and measure voltage somewhere else but I do need some pointers, please?
 Mar 10, 2015, 09:38 AM Registered User United States, CA, Norwalk Joined Apr 2004 2,746 Posts Okay, we must be missing something here. The obvious answer is to use a larger value pot so current is reduced, but there must be some reason you need to use this particular pot. If you want a 0 to 5 volt signal to measure, you'll have to have 5 volts across the pot, but you don't have to have it on all the time. You could turn the voltage to the pot on, measure pot position, then turn it off. The 1 watt power dissipation will be a short period of time, so heating wouldn't be an issue. You could add series resistance to reduce the voltage on the pot to perhaps 1 volt, then use an op amp circuit to bring the voltage back up to 5 volts. Dan
 Mar 10, 2015, 11:59 AM Hamburger Joined Jan 2010 918 Posts yes, I must use this particular pot. If I add another resistor in series, then I lose resolution, even on the analog side; right? But I need not measure a 5V input signal; only limit is that on the output side all I have for measurement is voltage reading in the [0; 5] Volts range. So I have - 5 Volt supply, - the darn 25 Ohms pot with 0,1W - a way to measure voltage range [ 0 ; 5 ] Volts. Need some simple magic.
 Mar 10, 2015, 12:12 PM "MAYONNAISE" Le Treport, France Joined Jun 2004 1,454 Posts except using an R/R OPA with a 3.5 to 4 minimum gain ... and feeding the pot with ~1.5v instead of 5 , I do not see any way ! and 25 Ohms is way too small to try an R-C analog/digital conversion ... Alain
 Mar 13, 2015, 06:46 PM Registered User Joined May 2013 83 Posts I would add a resistor on the high-side and then use a rail-to-rail I/O op-amp in the non-inverting configuration to create the final output. With a resistor of 250 ohms up to 5v, the current would drop to 18mA, which isn't bad, and the amplifier gain would only need to be set at 11. To get a gain of 11 with a non-inverting configuration, very simple values like 1k and 10k would be required. In the end, simple circuit, moderate power dissipation, and no loss of resolution. Martin Jay McKee
 Mar 14, 2015, 07:39 AM Hamburger Joined Jan 2010 918 Posts Hi Martin, sounds real good. I can possibly read with 1000Hz only, so the few first search results for rail-to-rail I/O op-amp do all by far outperform that frequency (no real surprise here). But I did not find any simple application diagrams for particular op-amps which would be clear enough for me how to set a desired gain. Could you please provide a pointer to such info? Thank you!
 Mar 14, 2015, 08:27 AM Registered User United States, TX, Van Alstyne Joined Dec 2014 210 Posts May I ask why a 25ohm pot is required in your application ?
Mar 14, 2015, 11:01 AM
Hamburger
Joined Jan 2010
918 Posts
Quote:
 Originally Posted by speedyb_tx May I ask why a 25ohm pot is required in your application ?
I try to bring back to life an old electromechanical controller and it so happens this one is built around a very non-standard 25 Ohm pot.
 Mar 14, 2015, 11:39 AM Registered User United States, TX, Van Alstyne Joined Dec 2014 210 Posts does the controller require the current or low resistance as presented by the 25Ohm pot ? any chance of a schematic (even partial) , made with express pcb ? (free program)
 Mar 15, 2015, 03:37 AM Hamburger Joined Jan 2010 918 Posts in original state, it is a power supply (about 24volts dc, 5A), the controller with the 25Ohm pot and the dc brushed motor (about 2A ?), all wired together in series. I want to convert the original electrics into an RC setup but preserve the physical controller look-n-feel. That is why I try to do with the given 25Ohm pot and convert with low current into digital. Last edited by hamburger; Mar 15, 2015 at 03:43 AM.
 Mar 15, 2015, 06:45 AM Registered User United States, TX, Van Alstyne Joined Dec 2014 210 Posts Does the pot's wiper arm go to the brushed motor ?
Mar 15, 2015, 07:35 AM
Hamburger
Joined Jan 2010
918 Posts
Quote:
 Originally Posted by speedyb_tx Does the pot's wiper arm go to the brushed motor ?
in the original setup, yes.
 Mar 15, 2015, 07:39 AM Registered User United States, TX, Van Alstyne Joined Dec 2014 210 Posts BTW, Mouser sells a 25Ohm Pot, 5Watts. http://www.mouser.com/ProductDetail/...CEErc12UVdk%3d As far as A/D sensing from a low impedance, just put a 1k or so resistor in series with the Atmel chip's pin. with an 8bit a/d you have .0196 volts/bit resolution if the reference is 5v. Might also play it safe and place a shunt 4.7v ~ 5.1V zener and a shunt reversed polarity diode on the chip side of the above mentionend resistor. ( Overvoltage clamp and negative polarity spike killer )
 Mar 15, 2015, 07:45 AM Registered User United States, TX, Van Alstyne Joined Dec 2014 210 Posts does the return/ground end of the pot go directly to the Batt negative ? even if thru the wiring of controller ? if it does, then a PNP transistor solution is possible.
Mar 15, 2015, 12:52 PM
Hamburger
Joined Jan 2010
918 Posts
Quote:
 Originally Posted by speedyb_tx does the return/ground end of the pot go directly to the Batt negative ? even if thru the wiring of controller ? if it does, then a PNP transistor solution is possible.
I can rewire the pot all I want. It is just the pot itself and its haptic experience I want to preserve.