A better BIP373 ignition driver ?

[jeffmarsh750]

Would this be a better choice than the Bosch BIP373 driver. Looks like it will handle 5 more amps of current.

http://www.digikey.com/product-search/e ... P3&x=0&y=0

Thanks, Jeff

[elaw]

That's an interesting device!

It's clearly made for the application, and as you indicated can handle more current than the BIP373.

The only downsides I can see are 1) it doesn't have built-in thermal protection and 2) being an IGBT you have to deal with gate charge. With MOSFETS and IGBTs, the gate has significant capacitance. So even though such a device may have a "logic-level gate", it's possible it won't work well when driven directly from a logic output like the BIP373 will. You might need a gate driver circuit that can source/sink higher current in order to turn the device on and off fast enough.

[racingmini_mtl]

I've been using the ISL9V5036, which is from the same family but with a higher current capability, in my kits for a long time and I haven't had any negative feedback. With a 1k gate resistor, they can be connected directly to the CPU pins. Any delay this may cause will be consistent and can be compensated for in the firmware and tables.

As mentioned, they lack the thermal protection but they can handle more current. So I wouldn't say they are better but simply a different option if you know what you're doing. They may be better for some application but universally so.

Jean

[jeffmarsh750]

I was hoping you would chime in Jean. I would like to build the highest possible current coil driver in hopes of driving the best possible coil for the most ammount of spark current for my Harley top fuel project. Was thinking about stepping the voltage up to 25 volts to the coil and with a .5 ohm rating would be about 25 amps at 50%dc. May even add a capacitor (what Nology sells)to the coil outputs and test that as well. Are these drivers components you can sell with your board ?
I am open to any Ideas
Thanks, Jeff

[racingmini_mtl]

Jeff,

I sell the ISL9V5030 with my ignition board driver and this is rated for a maximum of 46A. With your setup you would actually be beyond that for the duration of the dwell time which might be too long for reliable operation. And I would definitely not put these anywhere near the ECU to help reduce the risk of noise being an issue so you'd need to hose the board and components in their own case. You would also need to provide for a very generous heat sink because they would generate a lot of heat.

I have to say that I'm not an expert on ignition. I can recommend some proven solutions but I would not feel confident in recommending something for such a high performance and unique setup. And I'm definitely not equipped to test such a setup to help you determine the best configuration.

I know the components I use work in a mainstream setup and there is plenty of margin. I can provide the same components and board for you but there would not be much, if any, margin in your proposed setup so you'd have to rely on your own tests or get the opinion of someone with a better knowledge on the subject than me.

Jean

[jeffmarsh750]

Jean, maybe i am missing something, but the BIP373 is rated to 16 amps but the IGN1A coil (.5 ohm) is supposed to be using these and even the Quadspark and they are rated at 19amps 40%dc and I would assume 12 volts but most are up to 14 volts.
With yours being rated at more than double the current I would think I could double up the voltage on a heavy duty coil and actually drive at least 25 amps through it. I will test it to failure before we run it on the bike. Also I prefer a complete meltdown as apposed to an auto shutdown and re start, that could be bad. These will never run for longer than 2 minutes at a time and under only full power or hi current for 7 seconds max.
Thanks, Jeff

[phoenix3d]

Hi Jeff.. Just my 5 cents worth..

I have been running my MS on NGB8202NT4 Ignition drivers.. (SMD PArts though D2Pak)

Been running them with 330ohm drive resistors directly from the cpu.. (MS1)

HAD Absolutely no issues (Apart from incorrect settings that fried the track off the board but never harmed the driver.. ;-))

PS Also no thermal or overload shutdown.. But who cares.. ;-)

They are (Working from memory now) rated at 20A Continuos and 50A Pulsed at 25'C.

Run them on our 1024HP 7M-GTE drag car.. The drivers never even get warm....

heres the link for the Datasheet if you're interested..
http://pdf1.alldatasheet.com/datasheet- ... 02NT4.html

Cheers & Best of Luck!!

[jeffmarsh750]

Good info thanks. What about doubling up 2 BIP373's that should double the current without being too complicated.

[ekam99]

Good info thanks. What about doubling up 2 BIP373's that should double the current without being too complicated.

Please, clearly state your application design.
COP? distributor? # of cylinders? coil(s) you'd like to use? planned boost?

[jeffmarsh750]

I will test many things before we make a complete determination on what to use. The application is a 200ci V twin Harley running on 100 up percent nitro. They can make over 1000 hp from 2 cylinders. It is odd fire. The reason Megasquirt will be use I put another post on trying to do some sort of electronic controlled Magneto system but that does not look like it will work, and with odd fire motor even 2 regular magnetos can only run 1 plug at a time unless split with 2 coils and I think that is more of a compromise. We need a long burn time on the spark (CDI will not work) to light such a cold fuel. The engine currently has 2 spark plugs per cylinder but we may add 2 more. I believe the current systems in use (MSD supposedly 800mj) is not working at high RPM (4500) with the leaner mixture after fuel is being pulled out. They put cylinders out then it just stops making power. Some teams actually use the old Dyna 3 ohm coil and a basic trigger and they have set some of the fastest times. So i am pretty sure even an IGN-1A will be a major improvement. I just want the most spark current/volts that can be had.

I am thinking of doubling up batteries to 25+ volts to the coils and using 4 or 5 milliseconds from the MS to controll them. I know that jacking up the primary voltage will help with the output but the current also goes up as well. I have also seen that Nology sells capacitors to go on to the coils to help jack up the voltage when the field in the coils colapses (similar to a points type condensor) this I will test too. We have a high pressure test bench to try many digfferent coil combos and set ups.

Thanks again, Jeff

[racingmini_mtl]

Jean, maybe i am missing something, but the BIP373 is rated to 16 amps but the IGN1A coil (.5 ohm) is supposed to be using these and even the Quadspark and they are rated at 19amps 40%dc and I would assume 12 volts but most are up to 14 volts.
With yours being rated at more than double the current I would think I could double up the voltage on a heavy duty coil and actually drive at least 25 amps through it. I will test it to failure before we run it on the bike. Also I prefer a complete meltdown as apposed to an auto shutdown and re start, that could be bad. These will never run for longer than 2 minutes at a time and under only full power or hi current for 7 seconds max.
Thanks, Jeff

Jeff,

The problem is that the ratings for the different components do not always refer to the same thing. So the max could be an average or a maximum for a short pulse or something else. The rating for the ISL9V5036 is 46A for continuous use at 25C (31A at 110C). So that would mean that an average of 25A should be correct.

However, the datasheet does not mention a maximum for a pulse so an instantaneous 50A (0.5 Ohm at 25V) is outside the rating available. So I might be overly cautious but that is why I'm saying there is no margin. And I think that the rating for the BIP373 and quadspark also relies on the internal overcurrent and thermal protection to allow pushing to the limit.

But if you're going to test it in your specific setting and due to the fact you don't expect to be running for a long period of time, I think the ISL9V5036 is definitely an option worth exploring since it is specifically designed as an ignition driver. And if you do test it, I would be very interested in the results (actually, I'd be interested in the results no matter what you test).

Jean

[jeffmarsh750]

I need two of your board kits Jean so I will order from your site. Will test and report back.

[transaxel]

I always use this one

http://www.irf.com/product-info/datashe ... 14c40l.pdf

Never had problems or faults.

Greetings Gerald

[LAV1000]

Maybe this one:
http://www.digikey.com/product-search/e ... ageSize=25

A lot of IGBT's are used in frequency inverters wich drive elektrical motors.
An elektrical motor is just a set of coils.

Get an eletronics guy involved to check if this or an other IGBT might work.

Or just test them till they don't smoke anymore !!

[ekam99]

Maybe this one:
http://www.digikey.com/product-search/e ... ageSize=25

A lot of IGBT's are used in frequency inverters wich drive elektrical motors.
An elektrical motor is just a set of coils.

Get an eletronics guy involved to check if this or an other IGBT might work.

Or just test them till they don't smoke anymore !!

Don't use/suggest regular IGBT in ignition application. It seems they'll be dead in seconds. Not only the collector voltage has to be clamped but also the main spec for ignition igbt is switching energy has to be 150-300+ mJ.

[LAV1000]

I am thinking of doubling up batteries to 25+ volts to the coils and using 4 or 5 milliseconds from the MS to controll them. I know that jacking up the primary voltage will help with the output but the current also goes up as well. I have also seen that Nology sells capacitors to go on to the coils to help jack up the voltage when the field in the coils colapses (similar to a points type condensor) this I will test too. We have a high pressure test bench to try many digfferent coil combos and set ups.

Thanks again, Jeff

Have you noticed, Nology also sells "hotwires".
Or were you refering to this in the above post ?

[piledriver]

Note the capacitive wires or plugs (Nology wires or Pulstar plugs) will result in a very short but very hot discharge, not far off what a CD ignition ends up with, and probably not what you want. They DO work to some extent acting as sort of an impedance matching network for the coil,

I have been running the same set of iridium Pulstars for ~60K now, but I'm not making 5 HP/cu inch. You can play on a stocker or even "normal" hipo motor.

You are probably going to need something like multiple surface discharge plugs unless you want to do the glow plug thing the TF dragsters do by the end of the run.
IIRC the TF mags put out something like 50A. Each.

For electronic control you are going to need something like an inverter welding or plasma cut power supply in between.
(This has been done, played with a primitive setup myself but it kept letting its magic smoke out while consuming the plugs in a few seconds)

[old guy]

Take a look at this. Don't know anything about it, but looks interesting.
http://www.aquapulser.com/performance_i ... stems.html

[Matt Cramer]

We tested an experimental plasma system on our dyno once. It put out so much noise interference that it not only caused issues with the ECU, but we also had to rewire several sensors on the chassis dyno. They're an interesting idea, but I'm not sure if they are ready to make it out of the lab and onto the track just yet.

[prof315]

Also beware of Nology wires and the like. Every time I have dyno'd a car with Nology wires we lost 10-15 hp compared to "traditional" plug wires. I have never tried them with MS though.