Coil regulation related schematic diagrams and issues

General English language SECU-3 discussions here. Обсуждение общих вопросов SECU-3 на английском.

Moderator: STC

User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Coil regulation related schematic diagrams and issues

Post by Fred »

Again, current limiting will prevent excess current, and replace it with excess heat, which you must dissipate.

If you look at the DC resistance of most coils, you're looking at around 1 ohm, maybe a little lower, so if the device can support 20A and your battery is at 13, you're just going to fry the coil, before the current limit kicks in. Secondly, you NEED the current limit to be 10A or greater for proper end-of dwell, anyway. The result of these two conditions, is that you will burn coils, if your software sucks, or the user is an idiot, no matter what you do.

The condition of a partial burn should be handled in init code, checksumming itself at boot time.

The condition of a stupid user really shouldn't be handled at all, you're potentially compromising the unit for all the intelligent users in order to do it. A truly stupid user will always find some ingenious way to break your system, no matter what you do :-)

Basically, if he sets wrong values, it's not just the transistor that is suffering, and, if they are good ones, they will survive just fine provided that its below their limit (likely) and not current limiting and getting so hot that it melts the solder (likely), it's the coils that will burn. If the user is that stupid, they also deserve to blow up their device, as they really shouldn't be trying to home tune an engine anyway.

BTW, "Dwell time" not "Accumulation time" ;-)

Excuse my opinionated stance, but there is a lot of truth in it. Current limits are nice, but have their own inherent limitations as illustrated above.

Assume 15A limit (reasonable) and 14.4V and 4 channels, and stuck on from bad firmware burn, that's 864w you have to dissipate, think about that for a second. Not reasonable, is it?

A genuine reason to want current limiting is a failed coil. A single failed coil will be at the dc resistance of the coil, or lower, throughout the dwell period, potentially exceeding the max current. Protecting the device from a bad coil is good, because A its only during dwell anyway, a small fraction of the time B the user will be well aware that its only running on 3 cylinders or not running if its a dizzy, C because of A and B the device won't get a chance to get hot, or cook the entire interior of the car.

What value of current did you plan to use?

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
STC
LQFP144 - On Top Of The Game
Posts: 2420
Joined: Fri Oct 22, 2010 10:47 pm
Location: Ukraine, Kiev
Contact:

Re: Coil regulation related schematic diagrams and issues

Post by STC »

I did plan to use currents around 6-8A.
Usually 1 Ohm active resistance is actual only for old (mechanical ignition systems) coils. Active resistance of coils which I use vary from 0.4 to 0.6 Ohm. Most of IGBT transistors usually can handle 20A current.
Board voltage is 14.4V. Let Vce is 1.2V, 0.5V drops on wires. Then:
14.4 - 1.2 - 0.5 = 11.7(V)
I = 11.7 / 0.6 = 19.5(A)
I = 11.7 / 0.4 = 29.25(A)

Current limiting is obligatory, I think
Author of the SECU-3 project. SECU-3 Engine control unit / Ignition control system
SECU-3.org (Русский)
SECU-3.org (English)
SECU-3 Club ВКонтакте
SECU-3 EMS Project Facebook
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Coil regulation related schematic diagrams and issues

Post by Fred »

Post pictures of your heatsink! :-p
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Coil regulation related schematic diagrams and issues

Post by Fred »

STC wrote:I did plan to use currents around 6-8A.
Too low IMO. Crappy VB921s limit at 7 or so, and regularly provide insufficient dwell energy on many modern coils. BIP373 is 11.5A others are even higher.
Usually 1 Ohm active resistance is actual only for old (mechanical ignition systems) coils. Active resistance of coils which I use vary from 0.4 to 0.6 Ohm.
I was talking about modern laminated ones as found in millions of jap cars, 50% of which are imported into Russia these days, too! Some are down that low, though, yes.
Current limiting is obligatory, I think
You'd better implement current fold back limiting such that the heat isn't excessive, or your limit will cut off it's nose to spite it's face. (burn from heat instead of current).

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
STC
LQFP144 - On Top Of The Game
Posts: 2420
Joined: Fri Oct 22, 2010 10:47 pm
Location: Ukraine, Kiev
Contact:

Re: Coil regulation related schematic diagrams and issues

Post by STC »

I did enhance in coil regulation algorithm. Now it is being tested. New algorithm contains some prediction tricks which lead to better precision of pulses. Especially, it is important when engine starts.
Author of the SECU-3 project. SECU-3 Engine control unit / Ignition control system
SECU-3.org (Русский)
SECU-3.org (English)
SECU-3 Club ВКонтакте
SECU-3 EMS Project Facebook
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Coil regulation related schematic diagrams and issues

Post by Fred »

I just looked through this thread again and couldn't find what is meant by "coil regulation" - do you just mean dwell? Assuming yes...

Can you describe what you mean by enhance with respect to dwell algorithm, dwell for me is just a lookup from battery Voltage.

When you say better precision of pulses, do you mean the end angle/timing and therefore ignition point? As length of dwell is absolutely non-critical and it sounded like you were saying that you made dwell time more precise which would be pointless.

Lastly, can you describe the prediction tricks? I'm curious!

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
STC
LQFP144 - On Top Of The Game
Posts: 2420
Joined: Fri Oct 22, 2010 10:47 pm
Location: Ukraine, Kiev
Contact:

Re: Coil regulation related schematic diagrams and issues

Post by STC »

yes. It will be good if you help to clarify terminology, because I have problems translating some highly tailored terms to English.

By saying "coil regulation" I mean calculation of moment in time before ignition. During this period of time coil accumulates energy. Is it "dwell"?

Previous algorithm was very simple. Immediately after each ignition event I add to the current value of timer time of one ignition period minus accumulation time. Value is being loaded into MatchCompare channel of timer. At the startup of engine RPM of crankshaft is not stable and has a big dispersion. In this case accumulation time e.g = 2ms may disappear at all.
New algorithm is improved version of previos. It uses calculated value like in previous one only for basic prediction. Later, on each cog of cranck wheel predicted value is being corrected because I know inter-cog period at the beginning and know inter-cog period on each following cog. When between moment of accumulation beginning less than 2 cogs remain I load remaining delay into CompareMatch channel of timer. This is simple description.
Author of the SECU-3 project. SECU-3 Engine control unit / Ignition control system
SECU-3.org (Русский)
SECU-3.org (English)
SECU-3 Club ВКонтакте
SECU-3 EMS Project Facebook
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Coil regulation related schematic diagrams and issues

Post by Fred »

OK!

A coil is "dwelled" such that it builds up energy in order to later fire (exactly at, or immediately after the end of the dwell period). The period of time that the coil is switched on is called the dwell period. You dwell a coil. etc.

One thing to remember is that unless you have a smart coil with dwell limits built in, it is safe to over-dwell the coils during start up in order to get better and more accurate ignition timing without risking underdwelling or firing late to ensure sufficient dwell. Thus, for startup, and startup only, it could be beneficial to just switch it on at some fixed input tooth/event (what you're calling cogs) and only time the actual ignition end of the dwell accurately.

Just my 2c, feel free to ignore! I have no idea how your implementation is, and you do, so only you know if what I've said is relevant :-)

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
STC
LQFP144 - On Top Of The Game
Posts: 2420
Joined: Fri Oct 22, 2010 10:47 pm
Location: Ukraine, Kiev
Contact:

Re: Coil regulation related schematic diagrams and issues

Post by STC »

Fred. Is it correct to say "Dwell regulation" or "Dwell time regulation"?
Author of the SECU-3 project. SECU-3 Engine control unit / Ignition control system
SECU-3.org (Русский)
SECU-3.org (English)
SECU-3 Club ВКонтакте
SECU-3 EMS Project Facebook
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Coil regulation related schematic diagrams and issues

Post by Fred »

I don't understand what you mean by "regulation" however dwell is a period of time, so "dwell time" is a little superfluous in the general case. It's not really regulated, it's just set. You set the dwell at X and the coils stays on that long, and when you turn it off you get a spark, simple really. Can you give me more of a sentence? Then I can suggest the most appropriate form :-)
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
Post Reply