Jump to content

Richard Hill

Members
  • Posts

    319
  • Joined

  • Last visited

  • Days Won

    14

Posts posted by Richard Hill

  1. 3 hours ago, BRD said:

    this is a bugger as i just got 1 of these and half way thought fitting it, so it wont work the the Link g4+ then? 

    Yes it will work with the link, I have spent many hours in the last few weeks testing multiple units on our 2jz test bed.  What I have discovered is that with the the new firmware and a tweak of the settings its performance is very good.  Please see the list below for the settings I have found to work well :-

    Firmware 1.09

    setfilter15 

    setcanformat1

    setcanid950

     

    This will put it into link mode and average the samples (it does read very fast in default mode which can lead to some increased noise in the output) 

     

    With a filter setting of 15 it still has a good time constant whilst averaging out the outlying readings. 

     

    HTH, 

    Richard. 

     

  2. On 10/22/2019 at 1:11 PM, cj.surr said:

    in addition, my timing in that 320kpa cell was fairly aggressive considering that I have never attempted to dyno tune it at that level and never expected the ECU to see that cell. It is only about 1 deg less than at 280 kpa. I can be pretty sure that the failure was due to preignition (my best guess) or extreme knock. 

    We always hard code safety into the main ignition table by ramping ignition down quite hard above the targeted/tuned MAP values.  We also ramp fuel up quite a lot in these areas too.  This is before we add any other layers of engine safety.  Belt and braces.

    HTH,

     Richard. 

  3. Yes, the Atom has a similar boost control strategy to all the other ecus. It has 4 aux outputs any one of which can be used to drive a boost control valve.  The only limitation is the number of outputs, you may max out with fuel pump, tacho, boost control and idle valve.  We normally use the monsoon for 4 cylinder boost applications, just because you have more expansion capability, but the Atom will work fine. 

    HTH, 

     Richard. 

  4. A gp pwm works well driving temperature gauges.  Frequency isn't too important as they are generally quite well damped and so the needle won't fluctuate too much.  Set up a duty cycle table vs coolant temp.  What car is it? 

    HTH,

     Richard. 

  5. Yes and no.  The first signs of a problem might be with a hot engine under low revs, so if caught early there is potential to prevent serious damage. 

     If you calibrate oil pressure into the ecu then you could light the check engine light at a configurable pressure. 

     

    Also, both the 350z cars I have seen had the oil pressure output as per my post above.   

    Screenshot_20191021_211227.thumb.jpg.0bb68d31320e399391e74f0132b2aa67.jpg

  6. What is causing the "dramatic fuel overheating"?  We have never come across this problem and use return fuel systems regularly on many different cars up to 900 HP without any issues. 

    The first thing I would look at is putting a temperature sensor in the return line away from the engine to measure what the actual fuel temperature is.  Then you will know how big the problem is and work out how to mitigate against it.  A returnless system may be the way forward but will not solve the problem on its own (if your fuel temperatures are actually that high)

    Closed loop fuel pump control is not as simple to setup as a standard regulated return system and can cause more problems than it solves if not implemented correctly.  

     

    If it turns out that closed loop is the only way to solve the issue then there are a few solutions, the cheapest of which is probably an OEM controller such as a 9D372 which is used on a lot of Fords, Range Rovers and Jags.  You will also need a pressure sensor near the fuel rail to give the ECU feedback for its control system.  

    You could also implement an open loop system using these controllers which when set up correctly (with a standard regulator return fuel system) can dramatically reduce the amount of excess fuel pumped around the system.

    image.png.5c4c45ba734b1b5c60b953212357d83a.png

     

    HTH,

     Richard.

     

     

  7. I do believe that Link fulfill all of their advertised specifications as per their limited warranty statement in the current catalogue :-

     

    "For off-road use only, not intended for 

    highway vehicles." 

     

    Nearly all aftermarket ecus are also for track use only and not qualified for the road.  

  8. I have tried powering the ecu from the starter motor terminal, and never had any success.  It has always resulted in a big voltage drop caused by the huge transient currents drawn by the starter.  Running separate feeds for the ecu and fuel pump should minimise the voltage drops.  

    The situation is often compounded by small, race batteries which in my experience only add to the difficulties. 

  9. 7 hours ago, cj said:

    Try setting divider to 1000 on the can stream setup for "lambda 1". The spartan adv3 manual says this value is sent as lambda x 1000 on the wire.

    The current lambda value is clearly wrong - it only ever changes during initialisation, so looks like it might be reading a status field not a lambda value. Are you certain the controller is sending in default mode?

    That's not the issue, I have another unit working with the same stream as posted above.  Yes it's in default mode as the 3rd byte is clearly lambda temperature.  

×
×
  • Create New...