The ET Mod installation

[Cowboy Dave]

http://www.4wdsupplies.com.au/mn-triton-spv-industries-egr-resistor/

[liamk]

is a 6k8 resistor still the correct value for completly bypassing the thermistor?

[NowForThe5th]

is a 6k8 resistor still the correct value for completly bypassing the thermistor?

Read the following, Liam, just a few posts ^^^ up there.

I'm about to do the single resistor mod on the mn.. 6k8 across pin 1 and 2.. Yl and bl wires leaving it a fixed 6k8 resistance .. As I don't have a scan gauge it's hard to see if it is working correctly.. Is this the correct resistance for the mn ?

Seriously? Have you read this thread?

You are about to fit a resistor of randomly selected value, and don't have the equipment to test your work.

I have personally seen an ECU fried from such random tinkering, and it cost the bloke over $2k for a new one.

For the price of a big night out on the turps, you can have a tested solution with warranty backup.

[liamk]

Ive read this entire thread 5th, and ive already done the mod. The reason I ask is, I read in another thread someone saying now that 6k8 is the wrong value so was wanting to know if there is an agreement now on what the static value should be for bypassing the thermistor.

[Cowboy Dave]

It depends on how you define right/wrong. Also when you say that people are 'now' saying it is the wrong value, in fact some people said that from the very start, which is one of the reasons this thread became so contentious at the time.

It is a value that has worked for the people who first did it in an ML.

It is however a higher value than is necessary for both the ML and the MN, particularly the MN.

Some would say that being higher than necessary would make it less than ideal and therefore 'wrong'. I would be one of those people because I am a bit of a pedant when it comes to stuff like that.

Others would say well that value works out of the box for me so it is right enough. I can't dispute that decision when made by one person for their vehicle.

I would dispute it though if they represented to the forum at large that it was the perfect value for all.

I imagine others will argue it is semantics and on some vehicles in some cases they may be right.

In the end I took the easy way out and left it to the expert who knows more about these motors than I am ever likely to know.

If you've already done the mod and it is working well for you so far I think the question I would be asking instead might be, what are the possible adverse consequences, if any, of having the value too high? I don't actually know the answer to that question, although I think I did read it at one stage.

[mIwoo]

there is also more involved than just chucking a thermistor in tony et mod- for example... his is (?) (was) adjustable. he can also program it for different temperatures- for example- if you live in the middle of australia where its often mid 40 degrees you need a different value compared to someone that lives in snow for 9 months of the year...

[BillMcQuade]

Putting a higher resistance value on a circuit that may be outside of its designed load may (under the right conditions) cause damage to the circuit.

A through-hole resistor is not a component that is designed to work as a load bearing component, so taping one into a loom (where it also generates heat) is undesirable. A bamboo skewer support does not count as load bearing.

A resistor that suffers mechanical damage can sometimes suddenly increase in resistance, prior to going open circuit, causing a small high-voltage event (albeit at very low current). This may damage any IC in the circuit.

Some of the ICs on the circuit may be a lot easier to damage than you think. Sometimes, even a soldering iron placed onto the loom can introduce an undesirable stray current into the circuit. So if you are soldering in the resistor with the ECU connected, you are taking a huge gamble.

I can rattle off a few more.......

There also seems to be an insinuation by some, that people who are fitting the off-the-shelf modification are simpletons, who lack the skill to cut a wire and solder in a resistor, and that somehow fitting a 4 cent resistor to your $35k+ investment makes you more able/brave than us mere mortals.

I will come out and say it directly; I'm glad if the 6k8 is working for you, but at the end of the day, it is a randomly selected resistor that has been proven to be an incorrect value.

Surely it is time to lock this thread and be done with it?

[Naff]

[koshari]

Putting a higher resistance value on a circuit that may be outside of its designed load may (under the right conditions) cause damage to the circuit.

Absolute rubbish , from an electrical perspective the input circuit for the maf is a bog standard resistor divider network putting a voltage onto one of the legs of A to D and if i could be bothered doing the mesh analysis i could even tell you the value of both resistors as could you. the input operating range is 1 to 4 volts and if you go outside this range your ecu will simply throw under-range or overrange CELS. an open cct on the sensor will just take the op amp input voltage to line voltage where as a short will pull it down to near zero. the dropping and pullup resistors would have been chosen so as they cannot exceed their wattage rating. As an Engineer Bill you may want to go into the finer detaile as per why we generally pick inputs with high input inpedances and the merits of good impedance matching with respect to design when selecting input componantry.

From a vehicle performance perspective there is more variability, I personally put a large value in series with the sensor on my engine simply because it worked fine with no side effects, (ML 3.2 for the record) others with 2.5s and 3.2 variable vane PAJs have experienced run on, the common consensus these days appears to be a static value in place of the thermistor round 3.5K.

A through-hole resistor is not a component that is designed to work as a load bearing component, so taping one into a loom (where it also generates heat) is undesirable. A bamboo skewer support does not count as load bearing.

there are plenty of ways to secure a static resistor to protect it more than satisfatory from mechanical damage, personally i encapsulated the resistor in a piece of mastic lined heatshrink, if you wanted to be uber fussy you could even mount it on some vero board,

A resistor that suffers mechanical damage can sometimes suddenly increase in resistance, prior to going open circuit, causing a small high-voltage event (albeit at very low current). This may damage any IC in the circuit.

indeed, and by crossing the road you may get hit by a bus.
a high resistance joint in a plug can also produce the type of variation you describe, and guess what, The plug and play solution contains 5 extra connection points, so by that theory you would be at 5 time more risk using this approach,

Some of the ICs on the circuit may be a lot easier to damage than you think. Sometimes, even a soldering iron placed onto the loom can introduce an undesirable stray current into the circuit. So if you are soldering in the resistor with the ECU connected, you are taking a huge gamble.

oh gimme a break

I can rattle off a few more.......

please do , i would be interested in just much more frightening you can paint the simple procedure of adding a singe descreet component to a rester divider network, heaven forbid if we actually had to handle a component with soma actual real practical risk like a CMOS IC??

There also seems to be an insinuation by some, that people who are fitting the off-the-shelf modification are simpletons, who lack the skill to cut a wire and solder in a resistor, and that somehow fitting a 4 cent resistor to your $35k+ investment makes you more able/brave than us mere mortals.

this is the bit of your post where i actually have to agree with you, simple fact is if you dont know what you are doing, why are you doing it???

I will come out and say it directly; I'm glad if the 6k8 is working for you, but at the end of the day, it is a randomly selected resistor that has been proven to be an incorrect value.

again there are no right or wrong values, some work for some and not for others,

Surely it is time to lock this thread and be done with it?

good idea , the equivalent thread on the Paj forum contains nowhere near the sensationalism as this one.

[Stoneman]

^i agree

[BillMcQuade]

Although you make some technically sound arguments on the circuit impedance, I can't agree with your statement. Neither one of us knows the exact schematic of the ECU, and changing the circuit impedance may have undesirable effects under certain conditions. Note that the whole of my post is painting a worst case scenario. I have seen some very strange things happen when playing with PCBs. They are built to a price, not a standard. If you wanted to sit down over a beer one day, I'm sure we could both swap stories over how the laws of physics seem to be ignored by some circuits (don't even get me started on RF).

I'm sure that you have secured your resistor very well, and yes, you could also use a bit of prototyping board and pot it afterwards. The vast majority of people simply solder the resistor into the loom, so this point is still valid.

Once again, I am painting worst case scenario when the resistor gets damaged. After all, it's not like the cheap SMDs in the circuit are isolated by optocouplers People should be made aware of the cons as well as the pros, no matter how remote the chance. True, the plug in mod adds a lot more joints/points-of-failure, but I'm not advocating that either. In fact, there are things about the plug in option that I don't like, but the fact remains that the method of changing the circuit impedance in the plug'n play is a more desirable solution. There are ways to achieve the same outcome yourself, It's not exactly a state secret how the plug in option works. I also advocate checking the EGR and the butterfly signals after completing the mod, whichever mod method you choose. I bet 90% of people don't, so they go on blind faith that the mod is working as intended.

I've seen a whole batch of ICs damaged by being mounted with a faulty iron. Can, and does happen. I'm simply reminding people to unplug the ECU when working on the circuit. It's good working practice to do this, and it takes all of 2 minutes.

I'm not going to get into a mud slinging argument over this. The picture that I painted was all caveated with the word "may". True, the odds are slim that there will be damage, however, for the one poor bloke that does this, and fries his ECU, it would be a costly exercise. There is a lot of crowing in this thread about how easy the resistor mod is to perform, but it is also necessary (and fair) to highlight some of the risks, no matter how slim. You clearly have experience in electronics, and sometimes experience blinds us to the fact that people with no technical skill may not be aware of some of the pitfalls, or even correct basic techniques. I've been guilty of making this assumption, and it has been costly.

Ask yourself; if you were selling the resistor would you be prepared to warranty that installing it in the most basic form (ie; taped into the loom) will not cause any damage? I wouldn't, and that is the option that I bet most people took.

And now I've added more sensationalism, and I forgot to quote the text. Bugger.

[koshari]

Although you make some technically sound arguments on the circuit impedance, I can't agree with your statement. Neither one of us knows the exact schematic of the ECU, and changing the circuit impedance may have undesirable effects under certain conditions. Note that the whole of my post is painting a worst case scenario. I have seen some very strange things happen when playing with PCBs. They are built to a price, not a standard. If you wanted to sit down over a beer one day, I'm sure we could both swap stories over how the laws of physics seem to be ignored by some circuits (don't even get me started on RF).

again the input to the maf is a simple dc voltage input protected by a voltage divider. your fearmongering regarding changing the impedance of the cct applies equally to the popular vondor supplied solution being provided to achieve the same outcome (which to your creidit is acknowledged in your last post.) . whilst i agree with your comments regarding RF and AC circuitry around modern circuits as a whole this isnt the case in this scenario. Capacitive coupling and induction issues are hardly going to be relevant here. in practice neither one of us needs to know the full schematic of the ECU as we are only discussing the input for the MAF value, and the resistor divider network is documented in the Mitsu Nonclementure that has been published likely without the copyright holders permission on many forums including this one.

I'm sure that you have secured your resistor very well, and yes, you could also use a bit of prototyping board and pot it afterwards. The vast majority of people simply solder the resistor into the loom, so this point is still valid.

agreed and as i said if you dont know what your doing why do it. still working on a car which is isolated by the nature of the installation tends to negate any real risk of a mains powered soldering iron not having sufficient grounding, or others may use a butane or battery powered iron which virtually eliminates such issues.
that being said i chose to install mine with a removable inline bullet terminal arrangement.

Once again, I am painting worst case scenario when the resistor gets damaged. After all, it's not like the cheap SMDs in the circuit are isolated by optocouplers

we will keep optocoupler discussion for applications such as maintaining separation against different voltage grades in ccts such as switching power supply feedback signals to the pumping stage. having said that most optos i have seen fail were due to the open collector output being overloaded rather than the input side. massive digress i know.

Ask yourself; if you were selling the resistor would you be prepared to warranty that installing it in the most basic form (ie; taped into the loom) will not cause any damage? I wouldn't, and that is the option that I bet most people took.

well simple fact is i aint selling anything.

And now I've added more sensationalism, and I forgot to quote the text. Bugger.

so have i but hey that life.

If you wanted to sit down over a beer one day, I'm sure we could both swap stories over how the laws of physics seem to be ignored by some circuits

and maybe we could discuss the merits of beer brewed unter de Reinheitsgebot rather than synthetically brewed on such occasions or the worth of doing bode plots on our audio speaker cabling routes.

[BillMcQuade]

again the input to the maf is a simple dc voltage input protected by a voltage divider. your fearmongering regarding changing the impedance of the cct applies equally to the popular vondor supplied solution being provided to achieve the same outcome (which to your creidit is acknowledged in your last post.) . whilst i agree with your comments regarding RF and AC circuitry around modern circuits as a whole this isnt the case in this scenario. Capacitive coupling and induction issues are hardly going to be relevant here. in practice neither one of us needs to know the full schematic of the ECU as we are only discussing the input for the MAF value, and the resistor divider network is documented in the Mitsu Nonclementure that has been published likely without the copyright holders permission on many forums including this one.

I'm still not going to agree but at least we are getting down to the nitty gritty. Although I was going to admonish you for using emotive words like "fearmongering" I won't. Nor was it my intention to draw RF circuitry into the discussion, it was merely a device to illustrate how some circuits sometimes function outside of their designed parameters after something as simple as changing a brand of diode. I agree that either method changes the circuit impedance, however, I am still pedantic enough to want an optimal (and adjustable) value that I can rationalise as appropriate. I am still not convinced that significantly increasing the impedance will give trouble free motoring. As I keep stating, there may well be a rare situation (possibly an abnormal event during shutdown) that momentarily causes an issue. I feel that this is not an unreasonable suggestion. This may be a one-in-a-million event, but I wouldn't want it to be mine.

we will keep optocoupler discussion for applications such as maintaining separation against different voltage grades in ccts such as switching power supply feedback signals to the pumping stage. having said that most optos i have seen fail were due to the open collector output being overloaded rather than the input side. massive digress i know.

To add to the digression, maybe it's not a simple DC input. Maybe the actual processor input signal is 1VDC @ full scale? I don't know if the circuit incorporates any DC/DC conversion. If they are using a separate IC for modifying the signal to the main processor, maybe it operates at 3.3VDC, for example. I know quite a few applications where seemingly simple processes have been complicated by incorporation of an IC to interpret a signal prior to processing. Possibility of this cost-adding hardware being in the ECU? Slim, but I can't say for sure. You are correct on the most common failure mechanism for optocouplers, but there is always the exception. Usually I cop the exception

well simple fact is i aint selling anything.

No, but I reckon it's still the pub test when it comes to recommending stuff to people. I always feel that If I wouldn't warranty it, then I wouldn't recommend it. Yes, I am pedantic. I'm not selling anything either, but I am prepared to give credit to someone who goes to the expense and trouble of developing a product, and will stand behind it with warranty.

and maybe we could discuss the merits of beer brewed unter de Reinheitsgebot rather than synthetically brewed on such occasions or the worth of doing bode plots on our audio speaker cabling routes.

Reinheitsgebot oder Tod

Any discussion on phase or gain margin is always a delight, as is any discussion on the appropriate dopant for p-type layers on GaN RF transistors.

[koshari]

I don't know if the circuit incorporates any DC/DC conversion. If they are using a separate IC for modifying the signal to the main processor

may i suggest a little bit of square root extraction preprocessing the signal or some other F(x) function ?

Any discussion on phase or gain margin is always a deligh

personally i prefer to tweak the derivative values,

appropriate dopant for p-type layers on GaN RF transistors

cant say i spend a hell of a lit of time rolling my own PN junctions these days.

Reinheitsgebot oder Tod

well at least we are well and truly in the same ball park with that one. even more so with the amendment to include malt.

[Bazzer87]

I'm not sure why I am following this latest debate on this thread. It makes absolutely no sense to me. I understand 1 in every 10 words. But it's fascinating. Kinda like listening to two people argue in a foreign language hahaha

[koshari]

I'm not sure why I am following this latest debate on this thread. It makes absolutely no sense to me. I understand 1 in every 10 words. But it's fascinating. Kinda like listening to two people argue in a foreign language hahaha

fact is this is all you need to know,

it surpasses all other discussion.

http://www1.american.edu/ted/germbeer.htm

[BillMcQuade]

I'm not sure why I am following this latest debate on this thread. It makes absolutely no sense to me. I understand 1 in every 10 words. But it's fascinating. Kinda like listening to two people argue in a foreign language hahaha

Robust, yet friendly discussion is often the path to a mutually agreed truth.

[AnOldFart]

...Surely, ...SURELY... there has been -more- than -enough- 'crap' written in this thread already....

..IF.. you -know- what you're doing, working with static 'sensitive' electronics, then go ahead and install either, the two 12 K resistors (series/parallel) wired option, or, the single 6.8 K resistor wired (in lieu) of the AT1 Thermistor, option.
Don't, just wire a 6.8 K resistor 'in series' with the Thermistor, as it offers no 'high' temperature compensation and won't keep your EGR Valve closed, in high Summer temperatures.
Either way, you also need to fully accept that -if- you 'fry' your ECU in the process, then -you- wear it.

If you don't -know- what you're doing, then take my advice and buy Tony's ready made Plug-n-Play EGR Mod solution, with it's associated 'warranty' ie, if -it- subsequently 'frys' your ECU then I'm sure that Tony, will be happy to pay you for it....

[explorer.dave]

$150 for a plug'n'play solution + peace of mind, no brainer really

[koshari]

this thread is like a pendulum, every now and again it slows to a halt and needs a bit of a giddy up to get it going again.

[viking shippy]

It reminds me of a cowpat in a paddock that the flys haven't found yet...

[mad992]

It reminds me of a cowpat in a paddock that the flys haven't found yet...

[borngeek]

I'm not sure why I am following this latest debate on this thread. It makes absolutely no sense to me. I understand 1 in every 10 words. But it's fascinating. Kinda like listening to two people argue in a foreign language hahaha

fact is this is all you need to know,

it surpasses all other discussion.

http://www1.american.edu/ted/germbeer.htm

finally a post in here worth discussing.

[AnOldFart]

It reminds me of a cowpat in a paddock that the flys haven't found yet...

I would have said the -exact- opposite....

[hvac guy]

6.8 k in series is way off, anyway thats the static value if taking the sensor out of the system. 3.2k in series works well. Useing resistors with thermistors is not new and has worked on many control logics in my industry as setback or set forward for afterhours temperature settings. I think u guys are over thinking this, the ecu is not that sensitive.