Jump to content
SAU Community

Recommended Posts

Therre was a ca/sr ecu that changed between sequential and batch, but considering the rb20 ecu only has 2 injector drivers it is batch only. No sure about ignition tho.

Pretty sure ignition is because you have 6 ignitors, if it was batch they would save money and use 2 or 3.

  • Replies 63
  • Created
  • Last Reply

Top Posters In This Topic

Not enough memory I guess the extra timing calculations would require alot more code CPU used has limited outputs

Im actually a programmer and one of my job was writing software for old processors, namely the same cpu used in these early boards. The code for writing to 2 outputs vs 6 and doing sequential vs batch is basically identical, the cpu still runs fast enough to process that many events. I can't see the CPU being the limiting factor, would have been something else but f**ked if I know what.

Funny enough I'm a programmer aswell yeah there wouldn't be a heap of extra code there is a bit of room available then again I've never worked on these chips they are probably faster than I give them credit for

You guys dont have a clue.

You realise in going to the haltech you will be batch firing injectors and coils! The std ECU is fully sequential on both. It will do everything about 100times better than the haltech ever will.

http://www.fueltech....pdfs/HalE6K.pdf
8 injector drivers as standard equipment
There is a reason haltech doesn't support the earlier ECU's anymore is beacuse they are shit.
You dont think it has something to do with the fact that the e6k is near on 10 yrs old now and used old technology, how many companies do you know of that keep producing things for 10yr old hardware when there goal is to always develop newer technology? Edited by W0rp3D

A bunch of programmers here haha.

Not a lot of additional code would be required to implement fully sequential injection/ignition. You basically just need slightly more sophisticated CAS decoder subroutine and cylinder counter and assign calculated outputs to different output ports based on that counter (using it as either input variable or index on inj/ign subroutine call). It's just one more routine and a couple of additional RAM values.

Those old CPU's are not very fast, but fast enough to handle engine management (some of early factory ECU's have hardware cap of 8000 rpm - been tested by Nistune developers, don't remember if RB20 is one of them). The only reason for not using fully sequential management I can think of is CPU may not have enough output ports.

Multiplexing can be a bit tricky in real-time applications like engine management. Requires some cleverly arranged synchronization.

On the other hand, it depends on how exactly CAS/load-calc-output data flow is implemented.

By the way I'm pretty sure I've read somewhere that R32 RB26 uses fully sequential management.

Neo motors use 16-bit processor, different family and higher performance. R34's use Mitsubishi 7700 processors, earlier Skyline ECUs use some kind of Motorola 8-bit clone manufactured by JECS, if I'm not mistaken.

EDIT: I mean I'm agreeing with you :) I'm just explaining why full seq. mode is easier for later ECUs

Edited by Legionnaire

Neo motors use 16-bit processor, different family and higher performance. R34's use Mitsubishi 7700 processors, earlier Skyline ECUs use some kind of Motorola 8-bit clone manufactured by JECS, if I'm not mistaken.

Yep, much newer processor with more features.

Not necessarily so. The CPU itself may be the same/similar, but its peripheral devices can be very different - it's an ECU architecture thing. E.g. RB26 ECU may incorporate multiplexing/decoding logic you were talking about.

Sure, it is all speculation only. The best way to do it is to open ECU case and take a look what's in there. And assembly code would be handy also.

But I seriously think they are all sequential, otherwise they would use crank sync trigger instead of CAS.

By the way 16-bit processors are seriously faster than 8-bit in this particular case because all AFM values and load calculations use 16-bit arithmetic.

  • 4 weeks later...

Hey I was wondering what ecu would be better for my setup mods are

Td06h-20g 8cm housing

550 injectors

3 inch exhaust

35 mm turbosmart external wastegate 17 psi

Stainless high mount

32 gtr cooler

I know nistune will be cheaper but lets put price aside on this one

so which one did you go with?

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



  • Similar Content

  • Latest Posts

    • There's plenty of OEM steering arms that are bolted on. Not in the same fashion/orientation as that one, to be sure, but still. Examples of what I'm thinking of would use holes like the ones that have the downward facing studs on the GTR uprights (down the bottom end, under the driveshaft opening, near the lower balljoint) and bolt a steering arm on using only 2 bolts that would be somewhat similarly in shear as these you're complainig about. I reckon old Holdens did that, and I've never seen a broken one of those.
    • Let's be honest, most of the people designing parts like the above, aren't engineers. Sometimes they come from disciplines that gives them more qualitative feel for design than quantitive, however, plenty of them have just picked up a license to Fusion and started making things. And that's the honest part about the majority of these guys making parts like that, they don't have huge R&D teams and heaps of time or experience working out the numbers on it. Shit, most smaller teams that do have real engineers still roll with "yeah, it should be okay, and does the job, let's make them and just see"...   The smaller guys like KiwiCNC, aren't the likes of Bosch etc with proper engineering procedures, and oversights, and sign off. As such, it's why they can produce a product to market a lot quicker, but it always comes back to, question it all.   I'm still not a fan of that bolt on piece. Why not just machine it all in one go? With the right design it's possible. The only reason I can see is if they want different heights/length for the tie rod to bolt to. And if they have the cncs themselves,they can easily offer that exact feature, and just machine it all in one go. 
    • The roof is wrapped
    • This is how I last did this when I had a master cylinder fail and introduce air. Bleed before first stage, go oh shit through first stage, bleed at end of first stage, go oh shit through second stage, bleed at end of second stage, go oh shit through third stage, bleed at end of third stage, go oh shit through fourth stage, bleed at lunch, go oh shit through fifth stage, bleed at end of fifth stage, go oh shit through sixth stage....you get the idea. It did come good in the end. My Topdon scan tool can bleed the HY51 and V37, but it doesn't have a consult connector and I don't have an R34 to check that on. I think finding a tool in an Australian workshop other than Nissan that can bleed an R34 will be like rocking horse poo. No way will a generic ODB tool do it.
    • Hmm. Perhaps not the same engineers. The OE Nissan engineers did not forsee a future with spacers pushing the tie rod force application further away from the steering arm and creating that torque. The failures are happening since the advent of those things, and some 30 years after they designed the uprights. So latent casting deficiencies, 30+ yrs of wear and tear, + unexpected usage could quite easily = unforeseen failure. Meanwhile, the engineers who are designing the billet CNC or fabricated uprights are also designing, for the same parts makers, the correction tie rod ends. And they are designing and building these with motorsport (or, at the very least, the meth addled antics of drifters) in mind. So I would hope (in fact, I would expect) that their design work included the offset of that steering force. Doesn't mean that it is not totally valid to ask the question of them, before committing $$.
×
×
  • Create New...