Author Topic: My transistor ignition and tachometer project (TigTac) (Read 11161 times)

PStechPaul · « **on:** January 28, 2015, 08:15:06 AM »

I have finally started on this project, after much discussion elsewhere, and thought it should be in its own thread. I made a simple conventional ignition test board consisting of a 12V coil, points, and condenser, which I connected to a power supply, and was able to produce a spark in a spark plug. Here is a little video clip showing its operation - crude, but a start:

http://enginuitysystems.com/pix/TigTac_1707.AVI

I plan to make a cam for the points and have it turn with a crank, so that I can better simulate the conditions of a model engine. Then I'll make a better video and show more details.

Next step will be adding the simple electronic firing circuit as I showed in other threads. Here is the basic circuit and simulation:

Don1966 · « **Reply #1 on:** January 28, 2015, 06:08:10 PM »

That's great Paul glad to see you getting started with it. I will be following along with great interest.

Don

b.lindsey · « **Reply #2 on:** January 28, 2015, 06:22:57 PM »

Paul, will the tachometer part be separate or will this one circuit fulfill both purposes? I am particularly interested in a tach circuit I might could build, though I will leave the design to you guys who understand the sparky parts

Bill

CHP · « **Reply #3 on:** January 28, 2015, 08:47:50 PM »

Bill are you looking for an analog type RPM
or a digital one

b.lindsey · « **Reply #4 on:** January 29, 2015, 01:01:03 AM »

Most things seem to be digital these days. Back in my high school years when I was still fooling around with model airplanes...Heathkit (remember them?) came out with a simple optical tach but the RPM readings were shown on an analog meter as I recall. It had a switch so that you could change between a couple of ranges of RPM's. Wish I still had it. I think the kit was about $20 back then but it had a nice metal case and such.

Bill

CHP · « **Reply #5 on:** January 29, 2015, 01:13:24 AM »

BILL if your answer is to me please specified what you are looking for
cheeras

b.lindsey · « **Reply #6 on:** January 29, 2015, 01:21:41 AM »

CHP, my original comment was one of more general interest in Paul's circuit. I am not in serious need of a tach at the moment but if I was I would prefer it be digital. The thought of being able to make on as opposed to buying one has some appeal even if my expertise is far from circuitry design.

Bill

PStechPaul · « **Reply #7 on:** January 29, 2015, 01:34:37 AM »

This design should be able to function as a tach, spark generator, or both. It will be digital, but it is easy enough to add a filtered PWM output that can drive an analog meter, if desired. A fully analog tachometer operated from the points or cam position sensor is fairly easy to build, using a frequency-to-voltage converter.

Here is an inexpensive (under $2) IC that can be used as a tachometer as well as many other applications, and some circuit diagrams are supplied:

http://www.ti.com/lit/ds/symlink/lm2907-n.pdf
http://www.mouser.com/Search/Refine.aspx?Keyword=LM2907

Microchip has a more expensive ($8) device (TC9400) that can do both F-V and V-F conversions:

http://ww1.microchip.com/downloads/en/AppNotes/00795a.pdf
http://www.mouser.com/Search/Refine.aspx?Keyword=TC9400

Here is a simple circuit using a common 555 timer:
http://www.seekic.com/circuit_diagram/Automotive_Circuit/DISTRIBUTOR_POINT_TACHOMETER.html

Here's a compilation of many links for tachometers:

http://www.discovercircuits.com/T/tachometers.htm

Here's a very simple circuit using just one transistor:

http://josepino.com/circuits/?tachometer-circuit

HTH. Enjoy!

PS: Here is a sample of a digital version using a PIC18F452:

http://www.pyroelectro.com/tutorials/digital_tachometer_rpm/schematic.html

The choice of processor is rather arbitrary and only 12 I/O lines are needed, mostly to drive the LCD display. I plan to use a module that accepts the three I2C lines from the PIC so only one more is needed for measurement. I may also add a Bluetooth module to the RX/TX lines so that a handheld device or laptop can read and log the RPM data, and possibly also adjust timing parameters and other things.

PStechPaul · « **Reply #8 on:** February 03, 2015, 06:54:27 AM »

Here is an updated working schematic for the TigTac. I just finished a marathon session getting the PIC16F1825 to drive the I2C connection to the parallel LCD display adapter, but it is finally working

, although so far I only have it showing Hello on line 1 and Goodbye on line 2.

If you are interested, the files for the project are contained in this zipfile: http://enginuitysystems.com/files/TigTac/TigTac.zip

PStechPaul · « **Reply #9 on:** February 04, 2015, 06:37:34 AM »

Here is a video clip of the TigTac circuit showing dwell in mSec and RPM on a 16x2 LCD display. I am simulating the points closing and opening with a pushbutton switch on the PCB:

http://enginuitysystems.com/files/TigTac/TigTac_1709.AVI

I have updated the PIC files in the zipfile. I will be updating them as I go along:

http://enginuitysystems.com/files/TigTac/TigTac.zip

The I2C_PES.c and I2C_PES.h files are specific to the PIC16F1825 but may be modified for other PICs with a MSSP module for I2C. The LCD_PES.c and LCD_PES.h files work for the simple adapter I got for a couple dollars on eBay, and it should work for displays up to 20x4 and possibly 40x4.
http://www.ebay.com/itm/IIC-I2C-TWI-SPI-5VDIY-Serial-Interface-Board-Module-Port-Arduino-1602LCD-Display-/201097851956

90LX_Notch · « **Reply #10 on:** February 04, 2015, 10:38:09 AM »

Paul-
This is very cool. I wish I understood electronics; it is totally foreign to me.

-Bob

Don1966 · « **Reply #11 on:** February 04, 2015, 07:49:37 PM »

Nice job Paul and I do have a question. Did you make previsions to add a dwell and timing adjust? Like 10 turn pots to an analog input. It need to be flexable so you don't just have to program it to one particular engine.

Don

PStechPaul · « **Reply #12 on:** February 05, 2015, 01:32:16 AM »

I have not gotten that far, yet, but the finished product can have those adjustments. The dwell is really only needed for cam-actuated mechanical points, and all that is really needed is enough time to saturate the ignition coil primary, at which point it holds its maximum inductive energy and any longer just wastes power in its resistance and the external resistor (if needed). I have thought about this and there are a couple of ways to deal with it. Since it will have adjustable advance which will also vary with RPM, it will be necessary to sense one complete cycle of points closed and open to determine the RPM (and the time before next point opening). Then the software will need to wait until the time that corresponds to the next firing, which will be the time of one revolution less the advance time and the dwell time. At this point the coil will be energized for the dwell time, and then the primary will open at the advance time. As long as RPM does not change a whole lot between revolutions, this should work out pretty well.

The settings for dwell and advance might be done via Bluetooth or a display menu and one or more pushbuttons or a keypad. The advantage of Bluetooth or other communication protocol is the ability to do data logging, and perhaps make adjustments while the engine is running. I am a bit constrained by the 14 pins of this particular PIC device, but there are others with 20 and 28 pins that cost about the same, and there is really no need to conserve space.

For instance, the PIC18F14K22 has 20 pins and costs only about $2 (and I have a couple):
http://www.mouser.com/ProductDetail/Microchip-Technology/PIC18F14K22-I-P/?qs=sGAEpiMZZMsnNKdmhDfM1H%252bN4%252b417SNT

And the PIC16F1829 is the same family as the PIC16F1825 but has 20 pins and costs about $1.80 (and I also have some).
http://www.mouser.com/ProductDetail/Microchip-Technology/PIC16F1829-I-P/?qs=sGAEpiMZZMsyrc46HKMtR5TsJh015Vw1

The PIC18F24K22 is also about $2 and has 28 pins and many advanced features:
http://www.mouser.com/ProductDetail/Microchip-Technology/PIC18F24K22-I-SP/?qs=sGAEpiMZZMu9ReDVvI6ax%252bSchEKPtlU%252b65%2f8MwgJSsM%3d

I have some PIC18F2420 28 pin devices which are now a little more expensive but I would not need to order them.

Since the processor is so cheap I might as well use one of the larger ones and have plenty of elbow room for new peripherals and functions.

Don1966 · « **Reply #13 on:** February 05, 2015, 02:01:18 AM »

Agree bud you need to program around a chip with more then you need, in case you need it. One thing that you need to consider to, is the cost of making the project. To get builders interested it has to be at a reasonable price and user friendly to the person installing it. Having external adjustments to trim the timing and dwell makes it more user friendly.
Keep in mind that we want the advance to be triggered by say any given speeds for any engine. Having an external adjust to trim this would also help setting up an engines advance because no two are the same.

The use of blue tooth and display s good for setting up and testing but what of the cost?

Don

PStechPaul · « **Reply #14 on:** February 05, 2015, 03:01:35 AM »

You can get a Bluetooth HC-06 slave module for just $4 including shipping.
http://www.ebay.com/itm/HC-06-Slave-Bluetooth-Module-Wireless-Serial-Port-Module-Communicate-for-Arduino-/371221433944

The pins connect directly to the RX and TX lines of the processor, and pretty much any modern computer or tablet has a Bluetooth transceiver, or you can get a USB dongle for about $5. You can run a simple, free serial port terminal program like ProTerm or HyperTerm which comes with earlier versions of Windows. The PIC would need to listen to the incoming serial signals and interpret any commands that are sent from the terminal. And it can send data from the PIC indicating the RPM and dwell and other information. Instead of the terminal program, an application could be written with a nice GUI and controls for adjusting the parameters.

Thus the hardware for Bluetooth may be cheaper than a couple of trimpots and switches or keypad, and it could even eliminate the need for the LCD display. It's amazing how inexpensive these things are. Most of the expense is in the electronics design, programming, and making a PCB. Even that can be made for about $2-$5 in quantities of 20-50 pieces or more (from China, of course).

Stuart · « **Reply #15 on:** February 05, 2015, 07:51:16 AM »

Paul
You are doing a great project and your log is very informative although I am reasonable with electronics ( built a computor from descreat components befor the snicker z80 using the zilog Z80) I am not up with the pic and its programming , normal programming is ok C and swift .

The only comment I will make but do not take offence not every one uses windows an do not have a pc in the house eg I only have macs , and I do not intend buying windows to run in boot camp
Is there a BT app to talk to your device

Stuart

PStechPaul · « **Reply #16 on:** February 05, 2015, 08:21:36 AM »

I know next to nothing about Apple products but I'm sure they must have Bluetooth as well as some sort of serial terminal program.

Some links:
http://support.apple.com/kb/PH6560
https://discussions.apple.com/thread/5295159?start=0&tstart=0
https://itunes.apple.com/us/app/pterm-ssh-telnet-client-terminal/id287269552
http://todbot.com/blog/2006/02/23/howto-mac-os-x-bluetooth-serial-port/ (Looks like Apple MACs use something called Zterm)
http://soft.udm4.com/downloading/apple_bluetooth_serial_terminal_android/
http://www.macupdate.com/app/mac/31352/coolterm (Free download)
https://itunes.apple.com/us/app/get-console/id412067943 ($10 Apple app)

You can also use a serial terminal app for an iPhone or iPad or an Android hand-held or tablet device.

Also, I'm pretty sure that MACs have a Windows emulation mode, so if I wrote a Windows GUI you could still use it.

Stuart · « **Reply #17 on:** February 05, 2015, 12:36:24 PM »

Thanks Paul

I will take a look but as apple have changed a lot with 10.10.x I am not sure

As to the emulation mode that's not there in any form because they would have to use microsofts api and pay a licence

Yes you can run windows native ,no emulation under boot camp but then you have to by a full copy of windows , VMware and parallels will do it on top of OS X but again you have to buy windows

I will post a request for info on the apple dev forum ( private for Apple Devs) which I am for testing OS X beta seeds , and for dev of software , but nothing like that you do

PStechPaul · « **Reply #18 on:** February 08, 2015, 07:41:10 AM »

I have just finished building and preliminary testing of the transistor drive portion of the circuit. I used a BUJ403A NPN BJT which is rated 1200V and 6A, and I am driving it through a 10 ohm resistor from an MJE170 PNP transistor with a 200 ohm base resistor. I was able to get a pretty fat spark at 6 VDC and a probably adequate spark at 3 VDC. I have a 0.47 ohm power resistor in series with the 1.4 ohm coil, which limits the current to about 3 amps at 6 volts. For 12 volt operation, this would need to be about 3 ohms and would dissipate about 22 watts. For model engines, probably 6 volts would be perfect, but perhaps it would good to be able to use a single 3.3V LiFePO4 cell or 3.7V Li-Ion.

I'll put together a better demo and take some video later, hopefully with the TigTac circuit connected. Then it will just be a matter of arranging for some user I/O to set parameters and read RPM, which is mostly software.

When I first started testing, it worked a few times and then the coil current turned on solid. I thought the transistor or another component had shorted, but actually I had neglected to disconnect the points from the original demo, and somehow they were not initially making contact. A piece of tape on the points was all that was needed.

I also need to make the cam and crank to operate the points, but it might be better to rig up an inductive proximity sensor and/or an optical system. The conventional points have a rather strong spring, and I think it would be better to use an electronic sensor on small model engines. What do most of you use? If you could send me a spare ignition set-up, it would help me tailor it to that purpose. I'll have it set up and working at Cabin Fever, so you can see it there and retrieve your mechanisms there, or I can return them.

PStechPaul · « **Reply #19 on:** April 10, 2015, 10:33:30 AM »

I have finally finished building the TigTac into an enclosure about 7.5"W x 4.5"D x 2.5"H. It has a 6V SLA battery inside, with an on/off switch to a pair of binding posts. There is another set of binding posts for connecting the generator (0-30 VDC), and a 3 digit LED panel meter to read either battery voltage or generator voltage, as selected by a switch.

The TigTac circuit with the PIC16F1825 drives a 16 character by two-line LCD display, which shows dwell and RPM as measured by contact closure or an inductive proximity sensor, which plugs into a DIN connector (from an old PC keyboard). There is also a Bluetooth module which sends the information, as well as running time, to a terminal emulator program on a PC for display. It took me a couple of days to figure out why the Bluetooth serial connection was not working properly.

The circuit is supposed to provide a signal to the coil circuit to generate a spark with programmable delay or advance, but I have not yet implemented it. However, I do have the circuit mounted on a board and it can be manually triggered with a pushbutton or mechanical points, and possibly also with the proximity sensor. I have also added a 200M high voltage resistor and a 20k resistor to ground, which may be used to measure and observe the coil primary voltage waveform on a scope. This is a 10,000 to 1 ratio so it could even be used for the spark voltage itself. I may play with that later.

I'll have these "toys" with me at Cabin Fever, and I'll be happy to discuss the technical details. I'll even bring my laptop computer which can show the information from the TigTac, and I'll have the Microchip PicKit 3 so I can show how it can be used to customize and enhance the operation.

stevehuckss396 · « **Reply #20 on:** April 10, 2015, 11:59:08 AM »

Are there any pictures? did you use a mini coil or a conventional aoutomotive coil? How small is the total package? I like the auto advance feature and am curious how well it works.

PStechPaul · « **Reply #21 on:** April 10, 2015, 08:51:17 PM »

I'll be taking pictures of everything at Cabin Fever, but now I'm just trying to get everything done and put together well enough to take and show. I am using a conventional large automotive coil, but it should work with anything. The ignition advance is something that will need to be designed. If the points or sensor are set so that they close and open before the point of maximum advance, then the circuit can provide a programmable delay, but if they are set at TDC, then the circuit would have to read the RPM and then calculate when to charge the coil and then create the spark. Actually, that would be necessary in any case, unless there were a true position sensor to read the actual degrees independent of speed. The speed probably will not change a whole lot from one revolution to the next, except perhaps when starting. I don't have a model ICE to test it on, but maybe I'll get more info from people at CF, and possibly even maybe borrow one to use as I develop this device.

News:

Author Topic: My transistor ignition and tachometer project (TigTac) (Read 11161 times)