Matronics Email Lists :: View topic - LM3914 Bar Graph circuit question

Matronics Email Lists
Web Forum Interface to the Matronics Email Lists

Get Email Distribution Too! FAQ Search Memberlist Usergroups Register

Profile Log in to check your private messages Log in

LM3914 Bar Graph circuit question

Matronics Email Lists Forum Index -> AeroElectric-List

View previous topic :: View next topic

Author Message

mdnanwelch7(at)hotmail.co
Guest

Posted: Thu Dec 16, 2010 12:30 pm Post subject: LM3914 Bar Graph circuit question

Hi guys,

 

  I have recently built the bar graph circuit suggested by Joe Gores.  (This 

circuit incorporates the LM7805 5 volt regulator, and works off the Ray Allen

servo potentiometer)

 

  Anyway, I got the circuit built and it works great!!!  It lights the LEDs as I intended.

 

  In an unrelated circuit;

  I have a completely separate 5K ohm potentiometer that the engine uses for 

regulating fuel enrichment (via my manual adjustment).  The 5K Pot is in place 

of the stock coolant thermister.  By substituting my own 5K Pot in place of the 

non-adjustable thermister, I can regulate the 'cold engine' (high resistance) signal sent to the 

engine's ECU (computer), thereby making the engine think the engine is still cold,

and subsequently enriching the fuel/air mixture.

  This is a VERY common modification on the GEO/Suzuki auto website, for guys that

installed turbo chargers (like mine).

 

  My question is;  does anyone have a suggestion how I can use that LM3914/LM7805

circuit with my 5K potentiometer...and NOT affect the ohm reading it would have normally

sent to the ECU if the circuit wasn't there.

 

  (For information purposes; The 5K Pot has 3 tabs.  When turning the knob one 

direction, resistance between A & B goes from 0 to 5K, resistance from  A to C 

goes from 5K to 0.)



 Mike Welch

 		 	   		    [quote][b]

mdnanwelch7(at)hotmail.co
Guest

Posted: Thu Dec 16, 2010 12:30 pm Post subject: LM3914 Bar graph circuit question

Hi guys,

  

   I have recently built the bar graph circuit suggested by Joe Gores.  (This 

 circuit incorporates the LM7805 5 volt regulator, and works off the Ray Allen

 servo potentiometer)

  

   Anyway, I got the circuit built and it works great!!!  It lights the LEDs as I intended.

  

   In an unrelated circuit;

   I have a completely separate 5K ohm potentiometer that the engine uses for 

 regulating fuel enrichment (via my manual adjustment).  The 5K Pot is in place 

 of the stock coolant thermister.  By substituting my own 5K Pot in place of the 

 non-adjustable thermister, I can regulate the 'cold engine' (high resistance) signal sent to the 

 engine's ECU (computer), thereby making the engine think the engine is still cold,

 and subsequently enriching the fuel/air mixture.

   This is a VERY common modification on the GEO/Suzuki auto website, for guys that

 installed turbo chargers (like mine).

  

   My question is;  does anyone have a suggestion how I can use that LM3914/LM7805

 circuit with my 5K potentiometer...and NOT affect the ohm reading it would have normally

 sent to the ECU if the circuit wasn't there.

  

   (For information purposes; The 5K Pot has 3 tabs.  When turning the knob one 

 direction, resistance between A & B goes from 0 to 5K, resistance from  A to C 

 goes from 5K to 0.)

 		 	   		    [quote][b]

mikerv6a(at)ao-cs.com
Guest

Posted: Fri Dec 17, 2010 12:12 pm Post subject: LM3914 Bar Graph circuit question

Hi Mike



If you would like to ensure that there is no cross-effect between

your bargraph measurement and the engine control circuit, you

can use a dual potentiometer. This is like the potentiometer

that you have, but it has two sections instead of one, and the

two sections are operated by the same shaft. This approach gives

you a signal for the bargraph that will track the mixture-control

signal without an electrical connection between the two.



To find a dual potentiometer at Digikey, enter "dual potentiometer"

in the search box. Also check the "in stock" box. In the next

screen you can find 5K linear potentiometers. The better-quality

potentiometers are not exactly cheap...



Mike Linse

 	  Quote: 	
		  

 Hi guys,



   I have recently built the bar graph circuit suggested by Joe Gores.

 (This

 circuit incorporates the LM7805 5 volt regulator, and works off the Ray

 Allen

 servo potentiometer)



   Anyway, I got the circuit built and it works great!!!  It lights the

 LEDs as I intended.



   In an unrelated circuit;

   I have a completely separate 5K ohm potentiometer that the engine uses

 for

 regulating fuel enrichment (via my manual adjustment).  The 5K Pot is in

 place

 of the stock coolant thermister.  By substituting my own 5K Pot in place

 of the

 non-adjustable thermister, I can regulate the 'cold engine' (high

 resistance) signal sent to the

 engine's ECU (computer), thereby making the engine think the engine is

 still cold,

 and subsequently enriching the fuel/air mixture.

   This is a VERY common modification on the GEO/Suzuki auto website, for

 guys that

 installed turbo chargers (like mine).



   My question is;  does anyone have a suggestion how I can use that

 LM3914/LM7805

 circuit with my 5K potentiometer...and NOT affect the ohm reading it would

 have normally

 sent to the ECU if the circuit wasn't there.



   (For information purposes; The 5K Pot has 3 tabs.  When turning the knob

 one

 direction, resistance between A & B goes from 0 to 5K, resistance from  A

 to C

 goes from 5K to 0.)



 Mike Welch

etienne.phillips(at)gmail
Guest

Posted: Fri Dec 17, 2010 1:57 pm Post subject: LM3914 Bar Graph circuit question

Hi Mike(s)



Another point to ponder is that there are two types of potentiometers - linear and logarithmic. The most common dual pots are logarithmic, as they're used as volume controls in audio systems, where the logarithmic change in volume produces a perceived linear fading. 



The chances are good that for your fuel system, you will require a linear response. To make sure, and to check what type the exiting pot is, with all power off move the wiper to the middle of its range and check the resistance between the middle and unconnected outer pin. If it's 2.5k, then it's a linear pot. If it's 1k or 4k then it's logarithmic. 



The bar-graph circuit will need a linear pot to be useful. If needed, you might be able to find a log-lin dual pot if you scratch around enough.



Hope that helps!

Etienne



On 17 Dec 2010, at 9:30 PM, MikeRV6-A wrote:



 	  Quote: 	
		   

 

 Hi Mike

 

 If you would like to ensure that there is no cross-effect between

 your bargraph measurement and the engine control circuit, you

 can use a dual potentiometer. This is like the potentiometer

 that you have, but it has two sections instead of one, and the

 two sections are operated by the same shaft. This approach gives

 you a signal for the bargraph that will track the mixture-control

 signal without an electrical connection between the two.

 

 To find a dual potentiometer at Digikey, enter "dual potentiometer"

 in the search box. Also check the "in stock" box. In the next

 screen you can find 5K linear potentiometers. The better-quality

 potentiometers are not exactly cheap...

 

 Mike Linse

mdnanwelch7(at)hotmail.co
Guest

Posted: Fri Dec 17, 2010 4:26 pm Post subject: LM3914 Bar Graph circuit question

> you can use a dual potentiometer. 

 > Mike Linse

  

 Hi Mike,

  

   Even though the dual potentiometer is fairly expensive, it sure does accomplish 

 what I'm after in a simple, yet effective way.  

  

   I think I'll head in this direction, and while I'm getting stuff from Digikey, I can

 round up another LM3914 and LM7805.

  

   Thanks for the suggestion, Mike.  Also, Etienne, thanks for the pointers on the 

 differences between the two types of pots. 

  

 Mike Welch

  

   

 		 	   		    [quote][b]

nuckollsr

Joined: 24 Mar 2009
Posts: 95
Location: Medicine Lodge, KS

Posted: Fri Dec 17, 2010 4:40 pm Post subject: Re: LM3914 Bar Graph circuit question

I have a completely separate 5K ohm potentiometer that the engine uses for regulating fuel enrichment (via my manual adjustment). The 5K Pot is in place of the stock coolant thermister. By substituting my own 5K Pot in place of the non-adjustable thermister, I can regulate the 'cold engine' (high resistance) signal sent to the engine's ECU (computer), thereby making the engine think the engine is still cold, and subsequently enriching the fuel/air mixture.



Mike, since the thermistor is a 2-wire device, your thermistor emulator potentiometer is NOT being used as a potentiometer . . . but as a simple adjustable rheostat (i.e. series resistor). From your post I presume you have an interest in having a light-up bar graph that represents the potentiometer position over full range of normal ops.



I think this can be done but it's a bit 'fussier' than the trim-indicator task you've already mastered. How much of the pot's rotation is used for full range of operations? Is this a vernier adjustment that is used to trim engine performance during normal operations? Or is it a "set here for cold starting" and "set there for operation?" Do you have some min-max markings on the pot's knob. How many degrees of total potentiometer travel are represented by the min-max pot settings?



The input impedance to the LM3914 is VERY high. The data sheet says bias current into pin 5 is 100 NANO amps max, 25 nA typical. This kind of 'load' will have no observable effect on the engine computer's perceptions of potentiometer position.



I would guess that one end of the "adjustable thermistor" is wired to ground. The other goes to some current source within the ECU where voltage developed as an effect of that current is translated to some temperature value. I would also guess that the minimum voltage is not zero and the max voltage for normal operating range does not require full rotation of the pot.



This means that your bar graph needs to be "differential". I.e. one led illuminated at the operational low but not zero volts and all led's lit at max volts (what ever that is). Can you get some measurements off an installed system for min-max volts to be converted into a graphical display?



Bob . . .,

mdnanwelch7(at)hotmail.co
Guest

Posted: Sat Dec 18, 2010 7:02 am Post subject: LM3914 Bar Graph circuit question

Hi Bob,

  

   What an incredibly astute understanding of the thermister's role.  As far as I

 can gather, 99.99% of what you said is how I understood it, too.  IIRC, the

 coolant temp sensor is one wire, depending on the body of it for ground. I agree 

 the pot is being used basically a rheostat.



   I haven't started the engine yet with the new turbo installation.  For low(ish) 

 RPMs, the stock ECU is supposedly capable of proper A/F ratios up to 5 psi.

 (with an unaltered coolant temperature sensor setup)  I have a low-range

 pressure switch (2-7 psi) attached to the throttle body intake housing.  Once boost 

 goes over 4.5 psi, it clicks over to the modified portion, allowing me to 'dial in'

 my own F/A ratio, from information displayed by the EGT, and A/F ratio, and boost 

 gages.

   Since most airplane engine's run at extended periods of time at specific RPMs, 

 I would also expect that the rheostat settings would be 'set it, and forget it'.

  

   I have not fired up the engine yet.  I am in the very final stages of hooking up

 my overhead and front panel.  In fact, today, I will be installing the overhead panel,

 and set the front panel into its premade brackets.  All I do then is connect the two DB15

 wiring harnesses, and the two panels are done.  Hook up the various sendors, and

 I ought to be able to hear some engine noises pretty soon.

  

   I can't really say how much of the rheostat's range would be needed.  I'll do some

 internet searching and see if there are any 'posted' voltage outputs.  I would suspect 

 there is not, though.  But, I'll look.

  

   I think what I need to do is A) get the electrical things finished   B) start the engine

 and get the timing set correctly   C) begin the anaysis part where I can find out what

 the stock ECU can handle...and what I need to augment.

   Once I have a whole lot of data to know what I'm dealing with, then I think I can 

 answer some of your well though-out questions.

  

   With your description of the complexities of the whole process, I think before I build 

 any kind of LED indicator, I ought to get the engine going first, do some testing, and THEN 

 determine how best to display the rheostat's setting.

  

   Thanks for the help, Bob

  

 Mike Welch

  

  

 		 	   		    [quote][b]

user9253

Joined: 28 Mar 2008
Posts: 1921
Location: Riley TWP Michigan

Posted: Sun Dec 19, 2010 6:00 am Post subject: Re: LM3914 Bar Graph circuit question

I thought of a couple more options for setting the mixture control.  A precision multi-turn potentiometer could be used with a knob having a numbered scale.  This is the simplest solution with no other indicator required.

  Another option is to use a volt meter to monitor the control setting.  A cheap Harbor Freight meter could be used, or a panel mount meter like this:

http://www.murata-ps.com/data/meters/20lcd.pdf

available at Mouser, part number 580-DMS-20LCD-1-5-C, $32.  Or a double-throw switch could connect the voltage monitoring pin of the EFIS to either the aircraft electrical system or to the mixture control.

  I am not recommending one method over another.  It is a matter of personal preference.  If you get the plane flying first, then you might be having too much fun to spend time puttering with electronic projects.  

Joe

_________________
Joe Gores

mdnanwelch7(at)hotmail.co
Guest

Posted: Sun Dec 19, 2010 10:17 am Post subject: LM3914 Bar Graph circuit question

> I thought of a couple more options for setting the mixture control. A precision multi-turn potentiometer could be used with a knob having a numbered scale. 

 

Hi Joe,

  This is exactly what I have already.  I have a very high quality potentiometer 

(5K ohm, 10 turn, linear) with one of those numbered dials that look just like

a "combination lock".  The dial even has a little lever that locks the dial in place 

onto whatever setting it is turned to.  The dial I got from Mouser is also a 10-turn

unit, so whenever you rotate the dial from zero (at 0 ohms) all the way to 10 

(at 5K ohms), it's reasonably close to indicating 1K ohms for every 2 turns.

  

   Here it is; (yes, the dial cost me more than $30);

  

 http://www.mouser.com/ProductDetail/Honeywell-Clarostat/31612/?qs=sGAEpiMZZMuiwDVLTMm01WUt3B34cWdtXy3YwOP582s%3d

 

 

  What I really would like, in addition to the numbered dial, would be a visual

indicator, also.  With Mike Pense's suggestion about using the dual pot and 

combining it with that LM3914 bar graph, this certainly get's the job done!!

 

   > Another option is to use a volt meter to monitor the control setting. 

 

  However, this suggestion of the digital number display (DVM) offers another 

 very good alternative to LEDs.  I read the data sheet and I do like the idea of a 

digital reading of the pot.

 

  Would I be able to use my present 5K potentiometer with this DVM, or would 

I have to get one of those dual pots, like Mike Pense suggested?  I really

would like the reading to be simultaneous.....that is, the ECU and I see the same

reading at the same time, as opposed to switching back and forth by way 

of a DPDT, etc. switch.

 

  Could someone come up with a design for the DVM that reflect the pot's setting, 

and whether or not this design would need a single or dual pot?

 

Thanks for your help,

Mike Welch

 

 

 

 

 

 

 

 



 		 	   		    [quote][b]

user9253

Joined: 28 Mar 2008
Posts: 1921
Location: Riley TWP Michigan

Posted: Sun Dec 19, 2010 2:55 pm Post subject: Re: LM3914 Bar Graph circuit question

Mike,

If you decide to use this voltmeter,

http://www.murata-ps.com/data/meters/20lcd.pdf

I do not think that it will be necessary to use a dual pot for isolation.  The meter has an input impedance of 800K.  When connected in parallel with a 5K pot, the resistance of the circuit will drop about 1 percent maximum (probably less).  If it were necessary to measure an exact voltage, then a 1 percent error might be significant.  But in this application, you are not looking for an exact voltage measurement, only the ability to repeatably set the voltage where the engine will run best.  A voltmeter will certainly be more precise than a LED bar-graph.

  The first step is to measure the maximum voltage across the pot, then buy a meter with the correct part number for that voltage range.  Connecting the meter to the aircraft is easy.  Use a 7805 voltage regulator to power the meter pins 1 & 3, connect pins 7 & 8 together, and connect pins 11 & 12 to each side of the rheostat.  If you want a decimal point, connect pin 3 to pin 4 or 5 or 6.

I do not want to give Mike bad advice; so opposing views are welcome.

Joe

_________________
Joe Gores

nuckolls.bob(at)aeroelect
Guest

Posted: Mon Dec 20, 2010 7:17 am Post subject: LM3914 Bar Graph circuit question

At 04:55 PM 12/19/2010, you wrote:

 	  Quote: 	
		  



Mike,

If you decide to use this voltmeter,

http://www.murata-ps.com/data/meters/20lcd.pdf

I do not think that it will be necessary to use a dual pot for 

isolation.  The meter has an input impedance of 800K.


	


   The voltmeter is a good lick whether or not it's

   left in place permanently. The thermistor being

   "spoofed" has a temperature vs. resistance characteristic

   that probably going to yield some narrow range of

   readings for normal operating conditions. I'd guess

   replacing the thermistor with just a potentiometer

   will provide at best a very coarse adjustment

   and only a small percentage of the pot's travel

   will be used.  It would be useful to have a voltmeter

   in place to quantify the adjustment rage for

   normal operations and then tailor some form of

   circuitry to emulate that range with large, high

   resolution rotation of the pot. A ten-turn pot

   is probably overkill but it's a very practical

   design goal to have normal ops settings to use

   80 percent of the potentiometer's travel with

   say 10% head and foot-room at each end.



   Crafting the circuit is easy but we need data.

   How much current flows in the thermistor sense

   lead and what voltages exist across the thermistor

   a the extreme ends of the operating range?



   From the operating ergonomics perspective, it

   would be better for the final configuration

   not to have a voltmeter. If this is a mixture control

   then you probably have "feedback" for optimal

   settings already in terms of EGT, oxygen sensor,

   engine smoothness, rpm drops from peak, etc. etc.

   ADDING another instrument to be observed, perceived,

   pondered and reacted to is probably going the

   wrong direction . . .





   Bob . . .

mdnanwelch7(at)hotmail.co
Guest

Posted: Mon Dec 20, 2010 7:45 am Post subject: LM3914 Bar Graph circuit question

> But in this application, you are not looking for an exact voltage measurement, only the ability to repeatably set the voltage where the engine will run best. A voltmeter will certainly be more precise than a LED bar-graph.

 	  Quote: 	
		   The first step is to measure the maximum voltage across the pot, then buy a meter with the correct part number for that voltage range. Connecting the meter to the aircraft is easy. Use a 7805 voltage regulator to power the meter pins 1 & 3, connect pins 7 & 8 together, and connect pins 11 & 12 to each side of the rheostat. If you want a decimal point, connect pin 3 to pin 4 or 5 or 6.

 I do not want to give Mike bad advice; so opposing views are welcome.

 Joe


	


  Hi Joe,

  

   You are correct, I don't need to know a precise value of voltage, only a consistant

 one that I can duplicate fairly quickly.  I would suspect there may be 3-4 maximum

 readings, i.e. take-off, climb-out, X-C fast cruise.  

   I'd also guess most of the other RPM parameters would not even need any of the 

 "coolant temperature sensor mod" contributions.  With the stock ECU able to handle 

 up to approx. 5 PSI boost, the DVM is going to be ONLY high RPM/high engine power

 ranges, no CTS mod necessary.

  

   I don't understand what you mean when you say "first step measure max voltage

 across the pot" and then buy the right meter.  Do you mean; after it is installed on

 the plane, and in operation, and the engine is taking a reading of the pot?

   This electrical expertise is above my experience level.

  

   One more quick question;  I follow you with all the pin connections, except for the

 two pot connections (#11 &#12).  Would these be hooked up in parallel with the CTS?

  

 Thanks a million,    Mike Welch

mdnanwelch7(at)hotmail.co
Guest

Posted: Mon Dec 20, 2010 8:26 am Post subject: LM3914 Bar Graph circuit question

> From the operating ergonomics perspective, it

 	  Quote: 	
		   would be better for the final configuration

 not to have a voltmeter. If this is a mixture control

 then you probably have "feedback" for optimal

 settings already in terms of EGT, oxygen sensor,

 engine smoothness, rpm drops from peak, etc. etc.

 ADDING another instrument to be observed, perceived,

 pondered and reacted to is probably going the

 wrong direction . . .

 

 

 

 

 Bob . . . 

 
	


Hi Bob,

 

  You do bring up a VERY valid point re: more complexities!!  I do have a digital EGT unit/gage 

w/ alarm function, a digital boost gage/w alarm, a wideband 02 sensor and digital gage.

  I hadn't thought about it before, but rather than rely on any additional stuff, maybe I 

should consider using the alarm functions to enrichen the A/F ratio?

 

  In a high demand power situation (high boost), it is possible the EGT could register high

temps (being too lean).  Same thing goes for the boost gage.  Supposedly anything 5 psi

and below is all automatically taken care of by the stock ECU.  

 

  I think I should fire the engine up, do some measurements, and THEN figure out which 

way to proceed.  

 

  I am very close to being able to getting it running.  The panels are in the process of being 

installed, and then I just need a couple of days to complete the misc. wiring and odds & ends.



   Here is a link to the EGT & Boost gages I have;  

 http://www.auberins.com/index.php?main_page=product_info&cPath=5&products_id=19&zenid=0aefca50e3651c54f87b3f3b7bf09f10

  

   At his point, I think I need data before I should proceed.

  

 Thanks for your help,   Mike Welch

  

 



 		 	   		    [quote][b]

user9253

Joined: 28 Mar 2008
Posts: 1921
Location: Riley TWP Michigan

Posted: Mon Dec 20, 2010 9:30 am Post subject: Re: LM3914 Bar Graph circuit question

 	  Quote: 	
		  I don't understand what you mean when you say "first step measure max voltage across the pot" and then buy the right meter. 	


The panel mount voltmeter that I suggested does not have a user selected voltage range like some hand-held meters.  You have to buy the correct model number for the expected voltage range.  If the voltage across the rheostat is 150mv, the 200mv model is the one to buy, not the 20volt model.  Unless you have some documentation about the circuit parameters, it will be necessary to measure (with a portable meter) the voltage drop across the rheostat with the circuit in operation.  Then buy a panel mount meter.

You asked, " 	  Quote: 	
		  I follow you with all the pin connections, except for the

two pot connections (#11 &#12).ï¿½ Would these be hooked up in parallel with the CTS?" 	
  I am not sure what you mean by CTS, but the meter input pins 11 & 12 would connect in parallel with the rheostat.

  I assume that you purchased a 5K pot because others have used it with success.  If it turns out that the pot gives coarse adjustments that make fine tuning difficult, it might help to put a fixed resistor in parallel with the pot.

   Eventually when you get everything working to your satisfaction, and determine the optimum resistance values for takeoff, climb, and cruise, then fixed resistors and a selector switch could be substituted for the pot.

Joe

_________________
Joe Gores

Display posts from previous:

	Matronics Email Lists Forum Index -> AeroElectric-List	All times are GMT - 8 Hours
Page 1 of 1

Jump to:

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum