Author Topic: LM2917 tacho circuit  (Read 15604 times)

0 Members and 1 Guest are viewing this topic.

commanda

  • Hero Member
  • *****
  • Posts: 731
LM2917 tacho circuit
« on: August 22, 2004, 09:26:29 PM »
The circuit is in PDF format. The pinout is for the 14 pin package.

* Commanda_Tacho_Circuit.pdf (6.92 kB - downloaded 165 times.)

And here's a shot of the veroboard.

<img width="80%" src="http://www.otherpower.com/images/scimages/2006/tacho1.jpg">

My 0-1mA panel meter is currently calibrated for 1000 rpm. I am optimistic, aren't I?

What I need to do is scan the meter scale, photoshop it, and print it to sticky paper.
Oh yeah, you can pull an analog voltage from pin 5. Feed it to a comparator for a star-delta switch. Or an rpm input to a mppt controller.
Amanda
« Last Edit: February 13, 2021, 05:23:08 PM by SparWeb »

BT Humble

  • Sr. Member
  • ****
  • Posts: 475
Re: LM2917 tacho circuit
« Reply #1 on: August 23, 2004, 04:35:12 PM »
Wow - that was quick!
What are the chances of you showing me how to convert one of the multitude of 12V shunt regulator circuits to 24V? 
* shuntregulator18.pdf (41.85 kB - downloaded 50 times.)

Unfortunately I still haven't progressed beyond being a "cookbook" electronics type of bloke. :-(
In return, I can, err, say lots of nice things about you?  Pretty please? :-D
BTH

("Push tab A through slot B...")
« Last Edit: February 13, 2021, 05:23:27 PM by SparWeb »

commanda

  • Hero Member
  • *****
  • Posts: 731
Re: LM2917 tacho circuit
« Reply #2 on: August 23, 2004, 08:40:12 PM »
This one's too easy.


Pin 6 is a reference voltage, typically 7.15 volts according to the datasheet.


Pins 4 & 5 are the inputs to an op-amp. Pin 4 is tied to the reference voltage.


All you have to do is make the voltage on pin 5 equal the reference voltage when the pot is in it's mid-position, and the batteries are at the voltage where you want the shunt regulator to switch on. It's a simple voltage divider.


Quick math, the upper 3K resistor needs to be about 8K2.


Amanda

« Last Edit: August 23, 2004, 08:40:12 PM by commanda »

BT Humble

  • Sr. Member
  • ****
  • Posts: 475
Re: LM2917 tacho circuit
« Reply #3 on: August 24, 2004, 12:38:59 AM »
That just goes to show that you need to ask the question of the right person!


Is there a specific kind of NE555 that I should use?  The standard ones go Phut* at about 15V don't they?


Sorry, But I did say I was slow! ;-)


BTH

* Technical term

« Last Edit: August 24, 2004, 12:38:59 AM by BT Humble »

commanda

  • Hero Member
  • *****
  • Posts: 731
Re: LM2917 tacho circuit
« Reply #4 on: August 24, 2004, 03:21:28 AM »
They don't go Phut*.


All the valuable smoke escapes, that's why they don't work anymore. It's all a giant con. They're actually all powered by tiny little steam engines. James Watt was far more advanced than the history books give him credit for. That's my conspiracy theory, and I'm sticking to it.


555 timers are made by many manufacturers. Look up some data sheets. What I have at hand shows a TLC555 with a max supply of 18 volts.


Alternatively, feed the positive supply through a resistor, say 100 ohms, with a 15 volt zener to ground, and a nice big capacitor, couple hundred microfarad.


Amanda

« Last Edit: August 24, 2004, 03:21:28 AM by commanda »

BT Humble

  • Sr. Member
  • ****
  • Posts: 475
Re: LM2917 tacho circuit
« Reply #5 on: August 24, 2004, 05:15:36 PM »
Do you mean like this?<p>

13854-0

(Sorry about the width, but it's quite hard to read if I squish the image).<p>
Do the zeners over to the right hand side (22V and 18V) also need to be changed?<p>
I hope I've got this right, or I'll look quite foolish! ;-)  <p>
BTH
« Last Edit: February 13, 2021, 05:21:24 PM by SparWeb »

commanda

  • Hero Member
  • *****
  • Posts: 731
Re: LM2917 tacho circuit
« Reply #6 on: August 24, 2004, 05:35:02 PM »
The 15 volt zener on the left is upside down.

The 200 uF should be in parallel with the zener, not series.


The 18 volt zener can stay.


The 22 volt zener needs to be higher, maybe 30 volts.


Amanda

« Last Edit: August 24, 2004, 05:35:02 PM by commanda »

commanda

  • Hero Member
  • *****
  • Posts: 731
Re: LM2917 tacho circuit
« Reply #7 on: August 24, 2004, 06:19:12 PM »
One more thing. Very important.


The top of the 8K2 has to go to the junction of the diode and 100 ohm resistor, so it's sensing the battery voltage (minus the diode drop).


I'll do a proper calculation on the value of that 100 ohm resistor later.


Amanda

« Last Edit: August 24, 2004, 06:19:12 PM by commanda »

commanda

  • Hero Member
  • *****
  • Posts: 731
Re: LM2917 tacho circuit
« Reply #8 on: August 24, 2004, 07:06:09 PM »
LM723 standby current drain 4mA max.

NE555 15mA


Lets say 20mA total.


Vbatt max lets say 28 volt. We'll ignore the diode drop.


So the voltage across our dropping resistor is 28 - 15 = 13 volts.

The circuit needs 20mA. Lets make it 30mA for some overhead. The zener will suck the rest.


Now, our resistor = 13 volts / 30mA  = 433 ohms. Nearest preferred value is 390 or 470.


Our minimum wattage on the zener is 15 volts * 10mA = 150mW. So a 1/2 watt zener will be safe. Although, if you unplug the 723 & 555, the zener current will be 30mA, giving a power dissipation of 450mW.


If you use a cmos 555, it's supply current will be much lower. So the resistor can be bigger.


Amanda

« Last Edit: August 24, 2004, 07:06:09 PM by commanda »

BT Humble

  • Sr. Member
  • ****
  • Posts: 475
Re: LM2917 tacho circuit
« Reply #9 on: August 24, 2004, 08:07:49 PM »
Thanks SO very, very much! (You can see why I didn't know where to start?)


Now, for my part of the bargain, ahem


"That Amanda, she REALLY knows her stuff!  She's so talented, I mean, EVERYONE on this board should be singing her praises!  And she's good-looking too!  And she can cook!  And she's the best driver I've ever met!  And..."


;-)


BTH

« Last Edit: August 24, 2004, 08:07:49 PM by BT Humble »

BT Humble

  • Sr. Member
  • ****
  • Posts: 475
Re: LM2917 tacho circuit
« Reply #10 on: August 24, 2004, 08:40:52 PM »
For the benefit of anyone else who's interested in this circuit, I've updated the image file (http://www.otherpower.com/images/scimages/236/shuntregulator18_24V.gif) with Amanda's changes.  I'll also be trying it out Real Soon Now.


BTH

« Last Edit: August 24, 2004, 08:40:52 PM by BT Humble »

commanda

  • Hero Member
  • *****
  • Posts: 731
Re: LM2917 tacho circuit
« Reply #11 on: August 25, 2004, 02:56:14 AM »
Now that we've done all that, have you seen redrok's shunt regulator.<p>

« Last Edit: February 13, 2021, 05:22:09 PM by SparWeb »

BT Humble

  • Sr. Member
  • ****
  • Posts: 475
Re: LM2917 tacho circuit
« Reply #12 on: August 25, 2004, 04:28:32 PM »
It's very nice, but it's still for 12V and I'd have an equal amount of difficulty converting it to 24V (as in, I'd have no idea where to start!).


BTH

« Last Edit: August 25, 2004, 04:28:32 PM by BT Humble »

commanda

  • Hero Member
  • *****
  • Posts: 731
Re: LM2917 tacho circuit
« Reply #13 on: August 25, 2004, 04:55:02 PM »
It's the same principle.  Voltage divider to a reference. Can be solved with ohm's law.


Amanda

« Last Edit: August 25, 2004, 04:55:02 PM by commanda »

ghurd

  • Super Hero Member Plus
  • *******
  • Posts: 8059
Re: LM2917 tacho circuit
« Reply #14 on: March 15, 2006, 03:49:45 PM »
Hey BTH,


How did this work out for use with 24V?


What values worked for you at 24V, for 390 ohms, and 8K2 ohm resistors?


I am trying it for 48V. Working out the 48V resistance values.

Thanks,

G-

« Last Edit: March 15, 2006, 03:49:45 PM by ghurd »
www.ghurd.info<<<-----Information on my Controller

Nando

  • Hero Member
  • *****
  • Posts: 1058
Re: LM2917 tacho circuit
« Reply #15 on: July 11, 2006, 09:33:16 AM »
I came late to this thread.

It seems that you have forgotten the Zener diode across the power Mosfet, it needs change to a higher value OR TO BE REPLACED by a Diode, ANODE connected to the DRAIN of the MosFet ( connection to the Ballast resistor) and the CATHODE (band) connected to the Positive of the Battery.


RESISTOR : 8.2 K ohms


DIODE = 1N4004 TO 4007 or any diode at least 1 amp and 200 Volts breakdown voltage.


Preferable to have a 5 to 10 amps diode for better protection in times of fast ON/OFF cycling.


This way any inductance causing oscillations in the line will be clamped to the battery voltage level.


24 or 48 volts ballast controller


For 24 or 48 volts IS BEST to have an isolated buffered supply to the controller.


Disconnect the controller from the Zener and bias resistor and filter capacitor.


Connect the BASE of a NPN power transistor to the junction of the Zener and the bias reistor and filter capacitor, connect the COLLECTOR to the Positive of the battery bank and use the EMITTER as the source power for the controller and add another filter capacitor.


Voila, you have a ballast controller with ample input voltage range.


The power MosFet must have a Voltage breakdown of at least 1.4 times the top battery voltage -- 100 + volts is OK, Current as needed and defined by the load and should be at least 2 times the carrying load.


Voltage reference control


The LM723 Voltage regulator has a 7.15 volts internal reference, so the input voltage at point of control needs to be 7.15 volts


The circuit says use a 1 K ohms pot.


For some the setting may be a bit problematic, so to easy such adjustment:


24 volts battery MEANS the CONTROLLER should be operating at around 28.44 volts and


for 48 volts the same point MEANS to operate at around 56.88 volts .


The best way is to use current in the resistor chain to set the upper resistors.


The ckt shows a 1 k ohm pot and a 3.3 K ohms to ground LETS ASSUME THE POT IS SET TO CENTER AT TRIP POINT

So the net resistor value is 500 + 3300 ohms = 3800 ohms


7.15 volts / 3800 ohms = 1.8815 milliamps


FOR 24 VOLTS : 28.44 - 7.15 = 21.29 volts

21.29 volts / 1.8815 milliamps = 11.314 kilo-ohms


USE 10 KILO-OHMS + 1.2 KILO-OHMS IN SERIES.


FOR 48 VOLTS BEST SOLUTION


28.44 volts / 1.8815 milliamps = 15.114 kilo-ohms - TO BE ADDED IN SERIES TO THE 24 VOLTS STRING .

IF ONE BUILDS THE CIRCUIT for both battery banks, one can select 24 or 48 volts by placing a switch in parallel with the 15.114 Kilohms resistor.


PROBLEM for some : the accuracy or precision of the resistor may CAUSE the setting of the pot PROBLEMATIC AND NOT ATTAINABLE,


SOLUTION : REPLACE : the 10 KILO-OHMS + 1.2 KILO-OHMS IN SERIES as follows :


a 6.4 Kilo-ohms resistor with a 5 Kilo-ohms variable resistor in series, connecting the 6.4 kh to the battery, the opposite side to the brush of the pot, also , brush connected to the Clockwise terminal of the pot, the CCW of the pot to the upper terminal of the shunt setting pot ( 1 Koh).


This way one can attain the proper ranges with resistor values that maybe not precise.


Setting : with the power of set the 5 Koh pot + 6.4 Koh = 11.2 Kohms for initial adjustment.


I hope this helps


Nando

« Last Edit: July 11, 2006, 09:33:16 AM by Nando »