RE: New CDI - opensource project JMJ and Bigboat
 

Here is a photo of a DA-100. It looks like the sensor is somewhere around 40 degrees.

From a software perspective the advance curve table needs to be set up from 40 degrees for starting (I'm just guessing here) to say 12 degrees for full throttle. Let me put it another way. It seems to me that we need two software versions - one for singles and one for twins. The basic software would be the same, just different advance curves.

Good catch John.

AlZ

RE: New CDI - opensource project JMJ and Bigboat
 

Hello SparkyGSX and thanks for your help.........

this software below is nearly bang on the money apart from i couldn't see a re-position entree to change the crank pickup

But it has everything else and more than we need to make ours work

http://www.ignitech.cz/exe/dccdip.exe

Can you use this software and change it to suit our needs ?

there another couple like this again ive found from different company's

Regards
Brett

RE: New CDI - opensource project JMJ and Bigboat
 

Is this software to control their products only? I see they have alot of old versions in the download directory along with hookup info for wiring etc. Anyone know what chips are in it? I've never seen this advertised anywhere...just asking..

John

RE: New CDI - opensource project JMJ and Bigboat
 

I ran this software on my PC..seems to access my PIC programmer on com-2.
I am wondering if it is adaptable to act as a go-between for our basic software once the tables are established? It does seem to affect the chip I inserted, changing the program I had in it initially..Not sure if it is working cause I have no means of testing it other than to do a verify-read of the chip afterwards..Maybe we are onto something here..???

John

RE: New CDI - opensource project JMJ and Bigboat
 

this is what I got as I saved the info to a file after changing my settings-rpm-advance points etc..

IGNITECH_KEY 67

BIKE 0
RPM_ADV 700 1400 6000 9500 11300 11400 11500 11600 11700 11800 11900 12000 13000 14000 15000
ADVANCE 7 9 23 28 10 10 10 10 10 10 10 10 10 10 10
POCET_BODU 4
BASE_ADV 6
LIMITER 100 150 6
POLARITA 2
GEBER_NR 1
PILOT_LIGHT 12000
REV_OUT_PILOT_LIGHT 0
CLUTCH 200 1000
NO_READING 0
RPM 1 0
INPUT 0 0



Is this useable info? Sure seems like it could be kinda plugged right into our data tables etc. doesn't it?

John

RE: New CDI - opensource project JMJ and Bigboat
 

John im not sure what chip they use......wonder if they would tell us if you asked.....

there are a couple more programs that are very nice looking like this one, there like a 3 in 1 program :D

1, configure
2, read
3, write

Brett

RE: New CDI - opensource project JMJ and Bigboat
 

From what I can see, the source code for that microcontroller isn't available, which means we can't modify it. The application for the advance curve etc. might be usefull, but that's something we should look into when we have new, working code.

RE: New CDI - opensource project JMJ and Bigboat
 

I’ve been thinking about some recent postings to this thread and would like to add the following comments and observations: Reading back over the post it’s obvious (to me anyway) that I have confused engine timing with sensor placement. Especially with regard to starting engines with the spark advanced 30 degrees. I apologize to all for the confusion.
Let’s step back for a second and ask why timing advance is important. Most of the small engines (chain saws, leaf blowers, and lawn mowers) are “dumbed down” for safety and product liability reasons. There is usually no advance and we live with the reduced performance of the engine at the high end. Therefore the engine is factory set with a compromise ignition setting - that is reliable starting with no kickback and compromised high end performance. Since this forum is about converting these so called “chainsaw” engines to RC use, I will focus my comments along those lines.
The first thing we should define is what makes a good ignition timing advance curve for an RC engine. I put together the attached graph to show the range we want to operate at. The first thing you will notice about this graph is that it goes way down on the RPM scale. This is because we hand start our engines and we need to approximate the speed with which we “flip” our props. The other thing to notice is that the upper end max is around 10,000 RPM. We very seldom see this. Typically are high end will be around 8500 RPM. Finally you can see that we start our engines around 6-8 degrees BTDC and begin to advance the ignition when we get to fast idle (around 2000 RPM). We then advance the ignition smoothly until it gets to the magical 28-30 degrees BTDC (around 4200 RPM) and there it will stay.
Now in the past there have been all sorts of schemes to achieve this. There have been mechanical advance systems. I attached a poor quality photo of one such system on a converted (glow to ignition) engine. Some systems did away with the transition entirely and used two micro-switches (start and run). The Jump Start is a battery powered booster ignition for magneto equipped engines. It will provide a hot spark at hand crank RPM. If your carb is set correctly, it will allow hand starts!
Here is where the beauty of the micro-processor comes in. It is simple, reliable, no moving parts and we can program it to do anything we want. Oops! I said we. Unfortunately most model airplane guys can’t program in assembly language and that is where we need help.
In the past few postings we discussed sensor placement and the most common location is ~30 deg. BTDC. Based on our timing curve we can see this is where we want the timing to be at the high RPM range. But trying to start the engine at this advanced timing setting would be futile and/or dangerous. But our timing curve tells us that we want to advance the spark based on engine RPM. So how do we determine engine RPM when starting the engine?
It is important that we know the instantaneous RPM when starting the engine to avoid dangerous kick-back. This is the RPM that we flip our propellers. Therefore instead of measuring the interval between pulses from the Hall sensor, I would propose measuring the duration of the pulse to calculate the RPM. This would generate an instant RPM reading. If the processor sees zero or low RPM it will immediately retard the spark so that it will always start at ~6 degrees BTDC. As a sidebar, CH is offering a neat little tachometer that plugs between the Hall sensor and ignition, but I dont know if they are using pulse duration or interval duration.
A flat (boxer) twin 2 cycle does not require a distributor as it fires both cylinders at the same time. One is on the compression stroke and the other is on the exhaust stroke. So the software for these engines should be the same as for single cylinder engines with respect to timing.
If my previous posts caused confusion, I apologize to all. If this is old hat to some of you, my apologies as well. Sometimes the only way to clearly convey a thought or idea is to be a little long winded. Also, for somebody new reading this thread, the information may be of help.

AlZ
Comments anyone?

RE: New CDI - opensource project JMJ and Bigboat
 

@Azalner: it's always good to make sure we're all on the same page.

I've thought about measuring the length of the pulse, but the problem I see there, is that very small variation in the distance between the crankshaft magnet and the hall sensor can make a very large difference in the pulse length, while this variation would be barely noticable while the engine is running.

I'm a bit worried about any system to delay or retard the spark when starting the engine, because we don't have any kind of reliable measurement, and there are human fingers near the propeller. I'd far prefer to adjust the sensor so that no timing adjustment is needed when starting.

I agree with your statement about the boxer type 2 cilinders engines, but only when we're talking about 4-stroke, ofcourse. I'd expect 2-stroke engines to be popular, because of the better weight to power ratio?

Anyway, any time of engine that requires multiple ignition coils could be supported in the future, but probably not in the first release.

RE: New CDI - opensource project JMJ and Bigboat
 

For the most part, the twin cylinder 2 stroke I posted a picture of uses a twin secondary coil with a single primary winding that needs no special attention for timing etc...It operates as a single coil would and I have used it with a fixed timing sensor with nothing more than a CDI board. What I had done previously was to make a throttle coupled spark advance (mechanical) for the purpose of getting a reliable run..it works great. If however, I left it in full advance, it wouldn't start easily..that being said, you can see the need for a low degree of advance for starting and then going to full advance as required for top end running..Any way we achieve this would be fine.

The software I ran yesterday to see what it changes would be ideal when we get our software tables established..but until then we are at a standstill. An RS-232 port added to the timer board accessing the data will work. We just need that type of interface for making changes..as things are now, the CDI portion functions great. Just my views for development of the system.

John

RE: New CDI - opensource project JMJ and Bigboat
 

I hope some-one knows the differend between 2 and 4 stroke engine tuning.

BTW, I stop with this topic/project.
PM's show me my comment is not appreciated.

RE: New CDI - opensource project JMJ and Bigboat
 

Rob,

What do you mean you stopped with this topic/project? This is your baby. Please don't give up now!

RE: New CDI - opensource project JMJ and Bigboat
 

azalner i dont understand you comments and your picture shows a 4 stroke motor with a 4 stroke graph,

A 2 stroke would need a curve like it power curve to get the most from it.....like Rob has shown in RED but i would have alittle more shape to it,

But he has shown the correct way to time a performance 2 stroke

Also Rob has said in the past this project is NOT just for you RC guys.....Its for all 2 stroke motors......so get off your hi horse and who ever is hasseling Rob by
PM leave him alone.... dick head

Brett

PS azalner if i miss understood your post then sorry

RE: New CDI - opensource project JMJ and Bigboat
 

Please resist the urge to curse, flame, degrade, insult or embarrass someone in your post. We encourage the free flow of your ideas, but believe that they can be communicated (and received) much more effectively if you keep things civil. If you have to vent, take it offline. We carefully monitor posts and will ban individuals who engage in offensive conduct within the forums. Thanks. (RCU Policies)

RE: New CDI - opensource project JMJ and Bigboat
 

azalner thanks for the reply.......you are quite correct....... But as you can see from Robs post he has had enough and without him we have no project

It would be cool if Rob had the respect hes deserved for sharing his ideas with us, no saying that most people dont......

please carry on, its a long thread and it is hard to read through it all but Rob would like it suited for all 2 stroke Motors as i said, and all ideas are good ideas


Cheers
Brett

RE: New CDI - opensource project JMJ and Bigboat
 

I think if Rob still wants to persue this then let him and the software guy do there thing. the spark is going to be customizable to suit any application. I don't know what the comments that were said,but I think we need to give Rob some space and stop badgering him in doing it the way we want and instead of letting him do it his way. Me included.

Rob Please don't stop because of some ignorant comments. You have a lot of time invested in this and it is really getting good. Please reconsider.,But if you decide its in your best interest to give it up then I do understand.

Charlie

RE: New CDI - opensource project JMJ and Bigboat
 

For sure.....but its to late charlie .......people only take so much......

RE: New CDI - opensource project JMJ and Bigboat
 

Yea I understand. So My next question is will SparkyGSX still help us? Please.........:(

RE: New CDI - opensource project JMJ and Bigboat
 

Charlie, are you still working on the timing table reversal for the original software? It would seem that most of us need to get that working first before changing to a new timer board design and chips. From what I've experienced with my LED timing light, the low rpm timing is still advanced too far for easy starting.. other than that, the curve seemed to be ok. Lower rpm starting point was the problem, am I correct?[:-]

John

RE: New CDI - opensource project JMJ and Bigboat
 

COM: don't worry, I don't give up this easily.

RE: New CDI - opensource project JMJ and Bigboat
 

Hi SparkyGSX,

thanks. That is very kind of you.

RE: New CDI - opensource project JMJ and Bigboat
 

yes SparkyGSX thanks so much for your time....it is very thoughtful of you

there is also this software from sportdevice it has the source code in C and asm plus there is a little program called Ignition Wizard all down the bottom of this page,
i have run the program before but crashed on me,( could have been my computer)

Is this what you need to make our's better etc

http://www.sportdevices.com/ignition/ignition.htm

thanks again
Brett

RE: New CDI - opensource project JMJ and Bigboat
 

Some questions / observations:

These designs only work (in the current form) with pickup coils, not with hall sensors.

They use the positive edge from the pickup the fire the SCR when starting, to get the proper point for the ignition.

I don't see how they interrupt the direct pulse from the pickup to the SCR in the CDI design.

I prefer some more over-voltage protection across the power devices, in particular the IGBT in the TDI design, and the MOSFET in the inverter for the CDI design.


The hardware design seems a solid, fairly simple and very reliable.

The software isn't quite what I'm planning to write, but it's simple and probably effective.

Has anyone tried one of these designs?

RE: New CDI - opensource project JMJ and Bigboat
 

Is the software at the link above useable for a 16f628? They are using a 16f84 which was what we had in the old timer board. Seems to me that if we switched from a 16f84 to the 16f628 without rewriting the code, we should be able to use the excel sheet they offer and the c-code they offer to make ours work.
Are their port designations in line with what the 16f628 chip is representing in the pinout or are there different inputs & outputs? If it is close, we could alter it for our use couldn't we? I would think that would be easier than rewriting a whole new code, since we already have CDI power boards working, all that seems to be needed is corrected code for our PIC chip. Am I wrong? Anyone have a different view on this?

John

RE: New CDI - opensource project JMJ and Bigboat
 

Hi SparkyGSX,

if you read the Notes below that CDi circuit it does tell you alittle bit how the circuit works, (didnt want to cut and paste)

it sounds like a low speed start circuit ?

John i think this sort of software looks cool and having it program our chip without the need for another program like MicroC or MBLAB is good also


Guys i have some other software i found awhile ago also, dont no if its any help for use

But it has some very nice features, You can drag the advance curve with your mouse and you can change the ''Advance sensor angle'' (in red)

i download the oldest version

http://electronic-ignition-system.co...i/download.asp

cheers
Brett