Can NanoEFI drive both AC and DC CDIs?

I noticed on the layout for the pcb, Your mag sensor is a one wire pin. On most larger bikes they have 2 wire pins for the cdi to have better timing control. The CDI uses the leading edge and trailing edge to calculate timing.

How does your CDI trigger work with both versions, If its only recreating half of the signal needed at the box?

Most two stroke and offroad CDI are wired differently that the system shown above.

I know this is for a simple engine swap, but the variety in ignition system for small engines is rather difficult. Kawasaki uses 2 pickups for 2 cylinders with 4 mag zones that determine the beginning and end of the stroke for either side.

What cdi are you including with the nanoEFI so I can plan ahead.

[Admin edit: Line spacing]

1 Like

I see in some of your post Travis that your unit will handle timing and save maps etc. but if you intend on using this with the factory cdi boxes of the engines in question, most of those systems already have timing maps built into the box, by way of resistors firing the transistors at the right moment. SO if that is the case then the signal produced by the NanoEFI will have to reproduce the factory signal for the factory CDI unit to work correctly?
I know I focus on ignition system, but I have learned through all my tinkering and modifying that having harmony between systems is key.
I understand that this is meant to be universal, but leaving piecemeal parts of the original system without some form of quality control or understanding of how the factory system works, it will be difficult for someone to adapt it. Including a simple CDI box, that is labeled AC or DC to work with the respective system, that your brain unit also controls exclusively will simplify the install for many.
The older bikes with points will need a complete ignition overhaul, and those with TCI type ignitions do not behave the came as CDI. TCI uses two pickups to have a load then fire signal. so that throws any TCI controller out the window. Regardless of two stroke or four stroke operation, I still believe that with advanced fuel control, the engine should have tight ignition control. Using a dumb type system that doesn’t account for vacuum load vs throttle blade position vs rpm will only create a moderate to good improvement. The more points of the fuel and timing control will give better economy, performance on deceleration (eliminate Decel pop when snapping the throttle plate closed) and give better “feeling” over the available power. A dumb FI system is only mildly better than a well tuned carburetor.

Hey @Shaggymech, I appreciate your enthusiasm but do me a favor a slow down juusssst a hair. I know you’re pumped and doing research fast, but much of what you’re posting is incorrect or inapplicable to NanoEFI. I’m a big quality over quantity man and prefer slower more deliberate conversation.

That said, I think you’ll be very happy with the answers:

A good question paired with an incorrect assumption :wink:

Not half… NanoEFI is able to generate both positive +12v and negative -12v sides of the trigger pulse just fine.

Further than that, you can can also set whether the generated pulse for the CDI is produced negative-first, or positive-first. It’s a toggle in Airtune’s GUI.

Speaking specifically to the GY6 AC and DC versions, both trigger exactly the same.

And also generally speaking, although there are many (many) variations out there, there is no “AC” specific trigger, or “DC” specific trigger. When you hear “AC vs. DC”, that’s referring to the power source. Not describing the shape of the trigger signal.

I’m not looking to officially support so many different types of CDIs. It’s just not practical. So my default answer will almost always be to “use the perfectly good and well tested CDI included in the kit”.

It’s best to let NanoEFI handle the intelligent bits of ignition timing, we just want a dumb reliable (and affordable) CDI to produce a spark when we tell it. Also it must be cheap enough to keep spares on hand, and readily available brand new.

I should say that it’s important not to confuse cheap with inadequate. NanoEFI has my name on it, and I won’t send out parts that need to be upgraded in order to work as expected. The CDI I’ve selected for NanoEFI is the original version GY6 DC CDI from 15 years ago. Before the design was iterated into the ground to reduce costs. The newer versions are the poorly performing nonsense you see all over eBay and Amazon these days. I have a lot of history with this unit both personally and professionally. It’s rock solid, priced exactly right, and sourced through my long standing supplier by the tens of thousands.

With that said though, just because I think a certain way doesn’t mean I’ll lock others into that paradigm if there is a practical way to keep other options open. So you’ll still be able to use other CDIs if you’re interested doing the legwork to dial it in. However, I’m not in a hurry to implement the CDI learning/mapping feature. I consider this beyond the scope of our initial release and it’s not a part of my plans for the near future.

Especially during BETA, I want everyone running as close to standard gear as possible. Doubly so when it comes to ignition and the potential for serious engine damage if guys are getting too experimental too soon.

That’s the plan. Only DC though. There aren’t any practical advantages to including an AC CDI option, but a lot of potential for compatibility headaches.

And I don’t want to manage more SKUs than necessary. Lean is king.

For updating most older systems, just use the included CDI and coil. You could call it an overhaul, but that conveys more hassle than it actually is to convert. Or don’t convert, and keep the points. Either way is fine.

The more critical issue is making sure to supply the ECU with an appropriate source of trigger timing from the crank. For simple installs NanoEFI supports sensorless trigger sources, even right off of the kill wire.

For more advanced installs, linear hall sensor input works right now with a bit of external circuitry. I still have decoder work to do for N-2 patterns, but this will be ready and buttoned up before BETA.


Use the CDI!

It’s important to avoid assuming that everyone has the same goal. The level of precision you’re asking for already exists in NanoEFI. But we should all be careful asserting a “right” way (especially myself). Everyone is going to have different expectations and only a certain amount of complexity they want to deal with. We should allow people to get as advanced as they feel ready to tackle, as they’re ready.

I prefer tight ignition control personally, and tend to spend too much time chasing zeros if I’m completely honest. However, for every one of us interested in nailing theoretical optimals, there are 9,999 others who straight and simple just want to ride each and every time they turn the key without fail. With minimal tinkering and the least wrench time possible. That requires that we loosen up on certain aspects of control so our tolerances are forgiving to the widest degree of skill levels, and variables in the field.

So in the end, you have the basic option to run sensorless triggering which allows for ignition timing of around ±1° and a painless installation. Or fabricate a 60-2 or similar encoder wheel setup and get sub-degree precision. The advanced features are there for those of us wanting them. And entry level functionality for the rest of us just wanting to kick that old smelly carb.

Speak for yourself man, I’m after a bit of healthy decel pop and gurgle :drooling_face:

In the end “to each their own” is what matters, my role here long term is to provide the feature set and flexibility to do both.

I would love to say i’ve done this much research in just a few days, but this is an accumulation of about 12 years of data and research. I built my first bike at 16. a 1975 Yamaha DT250 that was literally in pieces that I customized and tuned from a paltry 20 hp at best to a 45 hp beast. Using a modern PMA and CDI, and after tinkering with about 8 different carbs. Most of the FI research i have done was actually for the DT250, since that was my next option in trying to create a better mousetrap. So i am actually well versed in most ways to convert something. I am not a coder or programmer and that was where i stuck at in my quest for FI. Your project has been a backburner idea that has grown to become nearly an obsession. Especially after i built my RD350 race bike. The RD350 was a 75 hp lightning bolt that had limitations in fueling. That engine would have become alive with FI, I could never get enough fuel at WOT to keep the top end happy. even when using powerjetted 34mm TMX mikuni’s.
Thank you for the answers! My apologies for the nearly endless onslaught of questions and information, but I am trying to really understand the bones of your system compared to other systems that i know of.
Yours being an all new product has a lot of variation and things that will still be specific to how you design the system to work, which is why i am digging so deep for the correct information.
the reason i asked about the CDI type is that will dictate what kind of stator i will convert to. the XS650 and generally most older bikes, say 1980ish, run incompatible generators that will have to be changed to a newer design, and even then the gyo scooter uses a small flywheel, with a rather small taper that isn’t exactly the easiest thing to just mount on a crankshaft. Also the flywheel is mounted on the crank with an internally mounted stator. many bikes, quads, and side by sides use a stator mounted externally or on the cover and sits into the flywheel. which is radically difficult to adapt to.
The trigger is the same yes, but the cdi still either needs 12v+ with a step up transformer for DC stators or uses the source coil on the stator to load the capacitors of the CDI. Which makes the generator choice all the more crucial. I’ve made dumb ac type CDI’s with 25$ worth of a few electronics so that is the easy part yes, but being a perfectionist as well as an engineer, I try to completely understand the operation of the system down to its core. I am in no means confusing cheap with inadequate. I have built Jeep cherokee’s on a shoe string budget and junkyard parts that out performed full on tube bubbys. I was asking mainly about the CDI because I am very much of the same opinion to lose the TCI and go with a reliable and available replacements. I do again also apologize for any incorrect or incompatible information with your NanoEFI, but this is still a beta and their isn’t exactly a full installation guide to read and understand the system. That is what questions are for also.

Thanks for understanding and your support @Shaggymech :pray:

If it’s meant to charge a 12v battery, it’s compatible. We’re talking relatively low draw, only about 150mA (±50mA) for the ECU, and less than 1mA average consumed by the CDI at idle speed.

If you have any form of DC 12v charging system at all paired to a battery, you’re good. Keep in mind though, I’m strictly speaking to the ECU/CDI hypothetical you’ve posed.

The big power hogs are your fuel pump, and to a lesser extent the fuel injector(s). That’s why I speak of stator upgrades for some machines with weaker charging systems.

So power isn’t an issue for the ECU or ignition side of things. The more pressing concern for older bikes is having a good source of trigger. It doesn’t have to be clean thanks to our well filtered sensorless feature, but it does have to match a definable pattern that allows us to infer crank position.

There’s no need to convert to a GY6 stator if you have other options. It’s just my default recommendation due to its price and availability.

The step-up transformer (and all related circuitry) is built into DC CDIs. No problem there.

You don’t have to worry about stator coils. That’s all handled through your rectifier/regulator and battery.

If you mean that you’ll need to figure out an appropriate rectifier/regulator on bikes that don’t already have one, that’s correct. But not a big deal. There are plenty of inexpensive rectifier units that will work great. Nothing special needed.

To that end, I’ll carry single-phase and three-phase rectifier/regulators with official support & documentation as well. One less thing to worry about. I already have our two most likely candidates here on my desk. True to our core project values, they’ll be affordable and easy to get elsewhere online even if NanoEFI (the business) goes up in smoke. I don’t want to lock people into using proprietary parts.

No harm done, Q&A is why were here. If you have 100 questions, I say ask 101. My specific critique is about getting out ahead and speaking too confidently, or in absolutes, before getting an actual canonical answer from me with relevant details.

It’s fine to articulate a current or evolving personal understanding based on past experience, but be sure to frame it that way rather than as stated as matter of fact. Truth is, none of us are experts. If there really is such a thing. At best, we’re all continuously learning.

At least most of us, I hope.

On that note, facts related to NanoEFI’s development are fluid and also evolve over time, so I’m conservative with what I put out there to avoid ultimately misrepresenting the final released product. And I usually take my time responding anyway to make sure I’ve thought through and considered my answer well enough. So there is a notably slower pace to conversation here versus other online forums. Quality over quantity :100:

It’s nothing personal, I’m glad you’re here!

Again, thank you for clarifying that line of questioning. With that I am only concerned with my charging system and a relevant trigger system. If you would refer to my build thread I have many ideas laid out and exploratory thoughts. But that makes this much simpler on my end. I have an engine pretty much ready for your electronics on my test bench now.

1 Like