Julian Edgar, 50, has been writing about car modification and automotive technology for nearly 25 years. He has owned cars with two, three, four, five, six and eight cylinders; single turbo, twin turbo, supercharged, diesel and hybrid electric drivelines. He lists his transport interests as turbocharging, aerodynamics, suspension design and human-powered vehicles.
Beware the expert…
Experts? I wonder…
A long time ago I came up with an electronic way of overcoming a boost cut on a turbo car. For less than a dollar, you could prevent a MAP sensor output from rising above a certain voltage. And for that single dollar, the voltage level was also adjustable.
At that time (and remember, it’s long ago – when I was certainly a lot more innocent), I told a turbo workshop proprietor about the technique, and how I was writing a story on it for a national magazine.
As this guy sold commercially produced boost cut units for lots of money, I thought he’d be fascinated by the approach.
But no, not a bit of it.
“You shouldn’t put that sort of technology into the hands of people who don’t know what they’re doing,” he said.
“Today’s cars are complex and people will blow up engines.”
When I pointed out that someone could blow-up their engine if you gave them even a screwdriver, he was taken aback.
Of course, the reason he didn’t like the idea of a DIY $1 boost cut unit was that his market for $200 ones was likely to evaporate.
So one reason that experts hate amateurs working in their patch is that often it costs them money.
Another reason is more subtle: it implicitly belittles their expertise and training.
I remember once when I was getting a lot of arc welding done. I was building a supercharger bracket and the 10mm steel plate was being welded by a local welding specialist. He was doing a very good job, too – and at a high cost.
Watching him in action, I thought I should invest in an arc welder – and the next week I bought a cheap secondhand welder. When he was doing my next lot of welding I told him of my purchase, expecting that he would be enthusiastic.
But as with the boost cut man, again not a bit of it.
“It takes years of experience,” he said. “You can’t pick up the skill of welding just by buying one.”
I hadn’t even implied that I could, but again his negative vehemence was off-putting and surprising.
Given that I’d been pouring money into his pocket, I thought he might have said: “Great! If you need any tips or get stuck, just ask.”
But no, I’d hurt his pride by apparently suggesting than anyone could buy a welder and so be an expert.
The third category of expert to be wary of is the person who ridicules any simpler approach to solving problems at which they’re specialists.
I’ve seen this in action most recently, with my copper wire modelling approach to designing space-frame structures. The experts look at the approach and nearly choke on their cornflakes.
“You can’t work out any stress levels,” they say. “You cannot accurately model anything by that approach.”
Of course, when asked how else to design space-frames (without using engineering mathematics or complex and hard to access software), there are no answers. Apparently, it is better to simply not attempt to design and build a space-frame, rather than to do so with an approach that has huge benefits over using no aids at all.
The corollary of all of this is that experts believe you should work only in your professional field of expertise, that – in other words – no hobbies should be embraced, no amateur past-times partaken in.
When it’s put like this, you can see the mixture of vested interests, pride, and the inability to understand that expertise can be reducible, invariably results in expert denigration of non-experts attempting anything.
I’ve seen it so often over the years: not only the aforementioned boost cut over-ride, welding equipment and space-frame modelling; but also aerodynamic testing by wool-tufts and pressure measurement, intake flow testing by pressure measurement, turbo boost control using injector duty cycle as the sole input, turbo boost control using a pressure regulator, single dimensional voltage interceptors working on the airflow meter signal, making trial aero undertrays out of cheap plastic sheet or cardboard, making plastic intake ducts from stormwater pipe and fittings, forming door pods for speakers using expanded polystyrene, swapping-in springs from other cars, DIY detonation detection systems, fitting a rear anti-roll bar, using suspension on human-powered vehicles, fitting your own subwoofer, using a pot to shift a voltage signal, altering regen braking on a hybrid car – the list goes on and on.
In fact, I can state that for every innovative, DIY technique we have ever covered in AutoSpeed, at least some experts have suggested the approach would not work. But in every case the techniques have worked extremely well!
I reckon that often you can get further ahead by actively ignoring experts. That’s especially the case if the technique is one that you have devised yourself and have found to be effective…
on April 14th, 2009 at 12:58 am
I’m very much in agreement Julian on all counts, but it’s human nature and frankly speaking, something you should expect and not be surprised about. The vast majority of people have a narrow world-view. Best recognise this trait and exploit it, rather than being unstuck by it.
on April 14th, 2009 at 1:47 pm
I absolutely agree Julian. I’ve come across this many, many times when I’ve wanted to do, or suggest something, a little different to the norm, especially when I was younger. I think the reasons you give are spot on also. As a thinker ‘outside the square’ I’ve often struggled to reconcile this worldview (try it in a traditional workplace!) with what seems to be (to me at least) a ‘commonsense’ approach to solving a problem or getting something done.
My approach these days is to read lots first, then go and ask the experts questions or do a course etc, listen respectfully to what they have to say and then make my own decision. I’ll always be a fan of DIY though, it’s the reason I visit the website after all! And it puts some of the power back into the individuals hands.
on April 14th, 2009 at 5:18 pm
I suppose if it threatens their way of life, or they even feel it does, then you’re likely to get an negative reaction as Paris said, human nature. The creativity and enjoyment involved in coming up with your own DIY ideas or even using someone elses is extremely satisfying. It can’t hurt to have a heap of skills, right?
on April 14th, 2009 at 5:52 pm
Wonderful insight.
Old addage of get what you pay for, and I told you so’s sometimes will extract senses of man and woman alike to unkown bounds and circumstances.
Nobody is ever right, manage your own opinion and views, and try your best not to get screwed these days. Research is key. If you dont put any effort into it and expect to pay people to do it for you in unnaceptable, but damn the sexist properties of humankind, but there sure are beutiful people out there who help with thier last breath of life, to make it easy for you and us together. Trust yourself and your feelings, you can do it.
on April 15th, 2009 at 2:48 pm
While I totally agree with your observation that an expert is often nettled by an intrusion on their turf, I doubt if a blanket ignoring of experts is going to be very helpful.
An “expert” by definition is someone who has an in-depth understanding and knowledge of a topic. If you start to assume that your “common sense” always trumps expert knowledge you can sometimes find yourself on a slippery slope into the depths of all kinds of wacky pseudo science.
Rather than “actively ignoring” experts I’d suggest you simply challenge them to give a reason why it won’t work. If they appeal to their own authority (the space-frame modelling guys) or go off at a tangent (the boost cut guy) you know it’s time to put on your skeptical face and become an expert yourself.
on April 16th, 2009 at 11:52 am
According to the experts someone with no formal engine tuning training, but who has a fair idea of how engines work and has the right tools, can’t diagnose and fix driveability and cold start problems on a heavily modified car. Apparently…
One moved throttle bracket and a fixed vacuum leak later, the car did the same thing all the time. Four cold starts later (with the Haltech plugged into my laptop) and the car started every time when it was cold, didn’t randomly stall, and used noticeably less fuel.
Julian – Have you taken your stuff back to the experts to show them?
on April 16th, 2009 at 1:04 pm
I’ve long respected Julians views and ideas. From arguing over whether to paint a radiator in 96 on aus.cars, supporting the early days of free Autospeed, the change to subscription and his incredible approach to developing improvements he always has something to make me think.
As an experienced Engineer I find very much the same attitude from other so called experts as Julian has. Once they have found their niche area of competency they are hard to budge and refuse to acknowledge any improved methods. I’ve used them for their narrow field of expertise, but having been more of a problem solver in my career I have always taken some inspiration from Julians innovation and always look for a way of making everything better.
As for the space frame copper wire design method, I’m not sure what kinds of experts you discussed the method with but it is very similar to a fundamental engineering technique where a full diagram is drawn and joints numbered, free body diagrams analyzed and loads calculated. This gives loads in each member and in fact each member can be individually analyzed for stress. It’s basic and tedious but the only accurate way to resolve the forces without computers and FEA programs. Modeling with wire will give a better idea of the physical space requirements and is simply sensible practice.
on April 22nd, 2009 at 11:39 am
I agree too.
But I’m one of those very rare things these days _ a systems analyst_
Not not just an IT systems analyst but a ‘systems approach’ type.
an ‘expert generalist’.
And I do piss experts off repeatedly, usually by asking a question they don’t want asked.
Frankly I think reductivist/narrow expertise is one of the core problems for modern society.
Managers!? Spare me!
Here’s an example. What what is the single most common flaw or shortfall (error even?!) in ‘science’ when it is based in hypothesis testing?
I’ll wait a week.
Tim B
on April 23rd, 2009 at 8:01 am
surely that you’re testing the wrong hypothesis or made wrong assumptions before starting the test?
on April 26th, 2009 at 11:38 am
Julian actively ignoring experts will not help and is the exact same midset of the turbo workshop owner but only seen from your own point of view.
The turbo workshop owner is clearly not an expert, very sadly today there are very very few experts around but many people claming to be so particularly in the car performance industry.
That being said the $1 boost cut puts additional load on map sensors and can cause them to fail.
Making a boost cut/fuel cut deffender for $5 with an a buffer preventing load on the map sensor on the other hand works perfectly and is still very cheap.
The welder may be an expert but in this field they are trained from the beginning to not share what they know as this is what keeps them employed. It is a sad sad mind set but you need to keep this in mind when dealing with them.
As to your comment about all your DIY ideas have worked extremely well this one didn’t, “single dimensional voltage interceptors working on the airflow meter signal”
This design of yours doesn’t work suitably as you claim and it has been tested and proven to not be suitable. IMO this DIY design is a dud and the low cost cutting corners and not using an RPM input and RPM refference to the map adjustment has lead this device to be completely useless.
As to most of your other innovations such as storm water drain air intake, plastic undertrays these are all identicle to function to the design of those manufactured using scientific methods and extensive aero testing. So replicating exactly what is used by experts should and will funtion exactly the same which is why they work. Not exactly rocket science………
On a side note what you cannot test with the plastic undertrays is the detailed air flows to all areas. You can have other negative effects such as effecting the air flow through the radiator and engine bay. Ofcourse you can just wait and see if there are any overheating issues but it isn’t possible to take everything into account which is what an expert in the OEM design process would do. There is always a risk that something will not be quite 100% and will be stressed and fail prematurely overtime but this is the risk of modifying and the gains typically outway the risks.
Damian
on April 26th, 2009 at 11:58 am
As to your comment about all your DIY ideas have worked extremely well this one didn’t, “single dimensional voltage interceptors working on the airflow meter signal”
This design of yours doesn’t work suitably as you claim and it has been tested and proven to not be suitable. IMO this DIY design is a dud and the low cost cutting corners and not using an RPM input and RPM refference to the map adjustment has lead this device to be completely useless.
Absolute rubbish – you speak as if it wasn’t tested before it was released.
I personallly ran the unit on a Lexus LS400 V8 VVTi, a Nissan Maxima V6 turbo and a turbo’d Toyota Prius for literally years without a single problem.
With the unit fitted, the Maxima (a very modified car) even passed a roadside emissions test – and that included running zero oxy sensor feedback!
The turbo Prius had a custom made airflow meter using a much larger than standard internal diameter, and the unit allowed the engine to run perfectly.
If you think you need an rpm input with an airflow meter car, you simply don’t understand how management systems and engines work.
As to the udertray causing changes in engine bay airflow, you clearly haven’t read the articles we’ve done on the topic – in one case, that airflow change was exactly the intention, and was measured to be very effective.
on April 26th, 2009 at 3:57 pm
I would mildly disagree – you just haven’t had much exposure to “real” experts – the real ones are only too happy, and usually enthusiastic, to find someone who is interested in what they are doing. The biggest problem is Brain Overload as they try to give you so much information – remember at school, the one REALLY good teacher who inspired you , who you still remember? The hacks – totally forgettable, but with the same “expert” qualifications.
And to quote you re welding:-
“It takes years of experience,” he said. “You can’t pick up the skill of welding just by buying one.”
I hadn’t even implied that I could, but again his negative vehemence was off-putting and surprising.
Tis the manner rather than the matter of his speech – he is correct. Welding is as much an art as a science – it literally does take years of practice to be good. The rule I use is “what would happen if that weld broke?” , ie would anyone die/be injured. If the answer is yes, then I overdesign the thing to buggery, add extra brackets, bolts, whatever it takes. And use more welds than an “expert” would. So far, it works.
on April 29th, 2009 at 9:57 am
Julian misunderstands experts. He was asking them the wrong question. If you want a particular result, whether its a weld that won’t fail or a bridge that you absolutely want to sure won’t fall down when the traffic goes over it, ask an expert. You will get a method of doing it, whatever it is, that will work. That’s their job to know. But if you ask them will this new/untried method of get the desired result, of course thery will say, no you can’t be sure it will, don’t do it that way, stick to this proven way, they’re just doing the job of guaranteeing the result will be achieved because the way they say to do it proven and tested and debugged. But that doesn’t mean that there aren’t other ways of achieving the result.
What an expert says will work will work. But something an expert says won’t work might.
on April 29th, 2009 at 2:25 pm
“Absolute rubbish – you speak as if it wasn’t tested before it was released.
I personallly ran the unit on a Lexus LS400 V8 VVTi, a Nissan Maxima V6 turbo and a turbo’d Toyota Prius for literally years without a single problem.
The turbo Prius had a custom made airflow meter using a much larger than standard internal diameter, and the unit allowed the engine to run perfectly. “
This is the blub from Jaycars website which is also pretty much a word for word copy of the description of the kit.
Ref: High Performance Electronic Project for Cars – Silicon Chip Publications.
This unit is a huge revolution in DIY automotive performance. It gives you the power to completely tune the air/fuel ratio throughout the entire load range, at 128 load points, providing incredible mapping resolution and brilliant drivability. It uses the Handheld Digital Controller – KC5386, so there is no need for a laptop, and it supports both static and real-time mapping. It can be used on 0-5V and 0-12V signals, so it is compatible with all voltage output airflow meters and MAP sensors. It can also be set to work with 0-1V signals, allowing modification of EGO sensor signals. This unit has been extensively tested on a wide range of cars including Subaru Impreza WRX and STi, Nissan 200SX, BMW 735i, Lexus LS400, and Nissan Maxima. Kit supplied with PCB, machined case, and all electronic components. Kit requires the Handheld Digital Controller – KC5386.
You claim not only will it work suitably with AFM but that the kit can also work with Map sensors and EGO (exhaust gas oxygen sensors) to be a powerful tunning device and provide brilliant drivability.
Given the map sensor output doesn’t change with RPM at WOT particularly in a boosted applications without an RPM input this device is not suitable as you are adjusting the same load points at a number of different RPMs which potentially can be the same load point from 2000 to 7000rpm. From my testing result in the inability to correctly tune over the rpm range.
This device is also not suitable for adjusting narrow band oxygen sensors which this is advertised to do so. Within the kit there is a revision which states it cannot adjust closed loop tunning but perhaps worth informing Jaycar of this as there sales staff still to this day believe it can.
As to adjusting the LS400 AFM signal this kit unless the later model 1UZ is fitted with a different AFM the kit isn’t able to as the 1UZ-FE AFM signal is a frequency based signal. Please correct me if I am wrong as to the VVTi 1UZ AFM. If the kit is combined with a frequency to voltage converter on the input and then a voltage to frequency converter it can be used and I have also done this myself. This worked much better than attempting to adjust a MAP sensor output but again how well is debateable. With the 1UZ AFM there is still the same issue of the same load points at different RPM’s this was more of a problem with different throttle openings but it resulted in the tune not being acceptable.
“If you think you need an rpm input with an airflow meter car, you simply don’t understand how management systems and engines work.”
Just for reference I am well aware of how an AFM works and just to make sure this is clear AFM measure the volume of air going into the engine and some have a voltage output whilst others have a frequency output like the 1UZ AFM. The volume or air increases with RPM and hence the load points would increase unlike a MAP sensor output. This in principle at a first glance makes the single dimension adjustment possibly OK. But as it is still possible to have the same load point at different RPMs which may and does in almost all circumstances require a different adjustment. The end rest is a non ideal tune hence this really doesn’t work suitably. Given a large number of AFM use a frequency output instead of a voltage output this kit without addition of the FTV and VTF converts isn’t able to work with them.
“With the unit fitted, the Maxima (a very modified car) even passed a roadside emissions test – and that included running zero oxy sensor feedback! “
Please explain what is a road side emissions test? I am well aware of the EPA emissions tests in particular the tests conducted that are stated in ADR37 but I have never heard of this roadside test before. Are police are now carrying exhaust gas analysis equipment with them and perform tested while parked on the side of the road running the engine under no load??
“As to the udertray causing changes in engine bay airflow, you clearly haven’t read the articles we’ve done on the topic – in one case, that airflow change was exactly the intention, and was measured to be very effective.”
I probably haven’t read all your articles on aero testing and undertrays but I have read several of them. I was always impressed but for OEM applications I am aware of the level of design refinement and testing that goes on to take into account every possible consideration and situation. This level of testing is nearly impossible to reproduce and taking into account every aspec of a modification and the possible negative effects. Even for OEM with the extensive testing manufactures sometimes get it wrong. That being said again building an undertray that is very similar to undertrays that are factory fitted to some vehicles it is hard to go wrong. My reason for making a comment in this area is that I have seen a DIY undertray that ran into issues with heating and they were eventually removed.
Damian
on April 29th, 2009 at 2:41 pm
You’re taking me to task for what a shop chose to write about the product?!
I didn’t write their blurb and I am not responsible for the mistakes that they have made.
In fact, I looked up what the original book had to say, and I clearly state in the chapter on the DFA that the use of the DFA will be problematic on MAP sensors. This chapter is included as a reprint with every kit sold.
The LS400 I had (the last of the series – VVTi engine) used a voltage outputting airflow meter. Hell, do you think I just make things up? Do you think that the photo in the book showing the Lexus being tuned with the DFA is faked? Get real….
Mass airflow as measured by an airflow meter does NOT just vary with rpm. It varies with BOTH throttle position and rpm. This is a fundamental point you have missed, and allows the DFA to work extremely well on airflow meter signals.
In Queensland there are random roadside emissions tests with a sniffer. The Maxima passed one, even without the oxy sensor plugged in, using a DFA and a heavily modified management system.
If you haven’t even read our articles on undertrays, it’s a bit rich criticising them, isn’t it?
on April 29th, 2009 at 3:16 pm
“You’re taking me to task for what a shop chose to write about the product?! “
Jaycar have provided a near word for word description that matches the description on the DFA box certainly intended to be within the same context. IMO this is misguided information and needs a simple correction so that people are aware of what they are buying not just after they take it home and pull it out of the box to read the revisions.
“The LS400 I had (the last of the series – VVTi engine) used a voltage outputting airflow meter. Hell, do you think I just make things up? Do you think that the photo in the book showing the Lexus being tuned with the DFA is faked? Get real….”
Since when did I say you just made things up. I have not used or tested the 1uz VVTi AFM output which is why I said “Please correct me if I am wrong as to the VVTi 1UZ AFM.” I also pointed out with the FTV and VTF it is possible to adjust a frequency based AFM output.
“Mass airflow as measured by an airflow meter does NOT just vary with rpm. It varies with BOTH throttle position and rpm. This is a fundamental point you have missed, and allows the DFA to work extremely well on airflow meter signals.”
Arh no I haven’t missed this fundamental operating point, I am well aware of this point and this is exactly where the problem lies which can result in having the same load points at more than one RPM.
“In Queensland there are random roadside emissions tests with a sniffer. The Maxima passed one, even without the oxy sensor plugged in, using a DFA and a heavily modified management system. “
Very interesting I have not heard of this before. Can you please inform me of more details, how was the test done what did they use as a sniffer what was used as a reference??
“If you haven’t even read our articles on undertrays, it’s a bit rich criticising them, isn’t it?”
Julian do not attempt to miss represent me I have clearly stated that I have read most of them but I am not read every single auto speed article so it is possible I may not have read ever single aero article you have written. Also I wasn’t criticizing them hence my comment “I was always impressed” referring to your aero testing and undertray articles. I was just pointing out the dangers which is the same potential hazards of all modifications in line with your original comments.
on April 29th, 2009 at 3:27 pm
I don’t think it is worth pursuing this much further: I have tried the DFA on many airflow metered cars with great success. You can believe that or not believe that as you wish.
Here’s a question for you: name me one driving situation (any combination of throttle position and rpm) where the airflow meter output does not reflect actual engine load. There aren’t any.
In fact it’s very easy to demonstrate – just put a multimeter on the output of the airflow meter and drive around. Low voltage = low load, high voltage = high load, and all the voltages in between represent proportional load. Use high revs with low throttle angle, use high throttle angle at low revs – use whatever you like and watch the meter.
on April 29th, 2009 at 3:29 pm
In fact it’s rather ironic – you’re just the sort of ‘expert’ the column is about – has never tried it, but sure thinks he knows the answers!
on April 29th, 2009 at 5:30 pm
Can we see an example of your welding on a supercharger bracket julian. It’s no wonder the guy treated you lke an idiot. An amature welding 10mm plate with a cheap ARC welder.
on April 29th, 2009 at 5:42 pm
Ken, the expert did that welding with an inverter arc welder, and very good it was too. In fact, it was better than that – it was excellent.
on April 29th, 2009 at 9:15 pm
Since you haven’t answered any of my questions you ask a question in anger I can only assume you have had much criticism on the DFA.
“Here’s a question for you: name me one driving situation (any combination of throttle position and rpm) where the airflow meter output does not reflect actual engine load. There aren’t any. ”
I am not arguing this point.
“In fact it’s rather ironic – you’re just the sort of ‘expert’ the column is about – has never tried it, but sure thinks he knows the answers!”
Probably the most rediculous thing you have written. Clearly you didn’t read my comments in detail. “Never tried” it I have extensively tested the DFA on 3 cars using a MAP sensor, frequency output AFM sensor and voltage output AFM sensor.
Question for you, is it possible with an AFM to have the same load point at more than one RPM?
on April 29th, 2009 at 9:48 pm
Ok frozenpod (why do people use stupid names?), you said:
” I have extensively tested the DFA on 3 cars using a MAP sensor…”
I have just referred to the chapter in the book, from which the instructions for the kit are reproduced in totality, and I see wrote:
In a naturally aspirated car which uses a MAP sensor to determine fuelling, altering only high load mixtures may be difficult. This is because manifold vacuum will drop to zero when the throttle is fully open – irrespective of whether the revs are at 1500 or 6000 rpm. Modifying the voltage output signal of the MAP sensor will therefore lean the wide-open throttle mixtures right through the rev range. To avoid these situations, before you install the DFA, use a multimeter on the sensor to confirm that the signal changes in a way which is consistent with successful modification.
You say that you have extensively tested the DFA with a frequency outputting airflow meter. I see I wrote in that chapter (reproduced as instructions with the module you bought):
The DFA will work only with voltage signals – some airflow meters have frequency outputs, so the DFA won’t work with these meters.
Given that clearly you did not read, or did not understand, the instructions provided with the module, I’ll leave people to draw their own conclusions as to the likely veracity of your claim that the device did not work with a voltage outputting airflow meter….
I might also add that I do not remember ever receiving a single email suggesting that a correctly constructed and installed DFA did not work exactly as suggested in my original articles. We have certainly received emails from people who couldn’t get it correctly bench calibrated (ie there were construction problems), but none from anyone saying that, after it was correctly set up, there were problems.
on April 29th, 2009 at 10:02 pm
I’ll add something else. I think that kit was the most extensively tested electronic kit I have ever been associated with. In addition to the Prius, LS400 and Maxima, I also personally fitted and tested the kit on a Nissan 200SX, Impreza WRX and BMW 735i.
In every case it performed faultlessly, giving absolutely perfect driveability and the ability to seamlessly alter air/fuel ratios, either outside of closed loop, or in the case of the Maxima and BMW, across the full load range. In fact, I also got a professional tuner to drive the Lexus with it fitted, switching the DFA in and out on consecutive drives. That car was brilliantly well mapped from the factory (as you’d expect) and the driveability was identical with and without the DFA, but with the DFA fitted, the mixtures (as measured by a professional MoTeC air/fuel ratio meter) were tuned leaner at the top end.
I can say with absolute certainty that people having problems with the device either have built it incorrectly, are using it incorrectly (like on a frequency outputting airflow meter…) or do not understand its use (like on a MAP sensor).
on April 29th, 2009 at 10:19 pm
“In a naturally aspirated car which uses a MAP sensor to determine fuelling, altering only high load mixtures may be difficult. This is because manifold vacuum will drop to zero when the throttle is fully open – irrespective of whether the revs are at 1500 or 6000 rpm. Modifying the voltage output signal of the MAP sensor will therefore lean the wide-open throttle mixtures right through the rev range. To avoid these situations, before you install the DFA, use a multimeter on the sensor to confirm that the signal changes in a way which is consistent with successful modification. ”
The car was turbo charged not naturally aspirated and I have emailed you about the results and issues experienced by the tuner.
“You say that you have extensively tested the DFA with a frequency outputting airflow meter. I see I wrote in that chapter (reproduced as instructions with the module you bought):
The DFA will work only with voltage signals – some airflow meters have frequency outputs, so the DFA won’t work with these meters.
Given that clearly you did not read, or did not understand, the instructions provided with the module, I’ll leave people to draw their own conclusions as to the likely veracity of your claim that the device did not work with a voltage outputting airflow meter….”
Clearly you didn’t read my comments, re Frequency to Voltage Converter, turns the frequency AFM output signal into a voltage signal which can be adjusted by the the DFA then the output is fed into a voltage to frequency conversion. The FTV and VTF conversion works without issue……..
“I might also add that I do not remember ever receiving a single email suggetsing that a correctly constructed and installed DFA did not work exactly as suggested in my original articles. We have certainly received emails from people who couldn’t get it correctly bench calibrated (ie there were construction problems), but none from anyone saying that, after it was correctly set up, there were problems.”
IMO you are not telling the truth ie not just a slip of memory, I have emailed you in the passed (just after the DFA was first released and I was first testing it) and John Clark at a later date relating to the issues.
I highly doubt I have been the only person to email you with the issues.
on April 30th, 2009 at 7:51 am
Oh my gosh, I have been found out! The DFA didn’t really work on the Lexus, Maxima, Prius, BMW, 200SX and WRX! The photos that I have in that book of the tuning of the Maxima and the Lexus are faked! The results described in that book are all lies! The Maxima didn’t really pass a roadside emissions test while running the DFA, a huge airflow meter bypass and no oxy sensor feedback! The huge number of emailed complaints that I have received about the product have been exposed! The cars didn’t really drive perfectly yet still be able to have their mixtures tuned! A single channel interceptor cannot be used on airflow meter cars! Thank you so much “frozenpod” for showing me how confused I have been…. you have done me a huge service.
(Note to self: in the future, just ignore people like him.)
To put it absolutely as politely as I can, the issues you apparently have had are not ones that I have shared. Clearly, you were doing something wrong, whether in the kit’s construction or in its application.
on April 30th, 2009 at 12:16 pm
I think Clarkson had something to say about people called Ken….
Cheap welder, expensive welder, 2 pieces of fencing wire between a tree with a bridge to adjust current, wires directly off power lines through buckets of water to reduce current….you can still weld it as long as the welding rods are in good condition, and then there are the myriad types of welding rod and flux combination. Some that can punch well above their weight. The technique is to build the welds and maximise the penetration, as with welding 100mm thick RHA (rolled hardened armour) together to form a Tank. If you tried to conduct a single pass the energy required would be extreme and the damage to the properties of the steel extensive.
Julian, I’m so glad I didn’t concentrate on being an expert with my Engineering career mate. I would have simply regurgitated the same thing over and over again ad infinitum without ever having played with the wide variety of fields I have. I may even be the only Engineer to full size test an Armoured vehicle in a wind tunnel. Flawed, naturally due to a blockage ratio of over 1/3 but we focused on a particular area of flow and optimised to obtain a valid result. Autospeed has always fed my urge to come up with solutions to do things different, cheaper, better and faster.
on May 1st, 2009 at 2:19 pm
“Oh my gosh, I have been found out! The DFA didn’t really work on the Lexus, Maxima, Prius, BMW, 200SX and WRX! The photos that I have in that book of the tuning of the Maxima and the Lexus are faked! The results described in that book are all lies! The Maxima didn’t really pass a roadside emissions test while running the DFA, a huge airflow meter bypass and no oxy sensor feedback! The huge number of emailed complaints that I have received about the product have been exposed! The cars didn’t really drive perfectly yet still be able to have their mixtures tuned! A single channel interceptor cannot be used on airflow meter cars! Thank you so much “frozenpod” for showing me how confused I have been…. you have done me a huge service.”
When I have I said any of this was a lie…. I haven’t but as you have repeatedly failed to read what I have written you have made another ridiculous statement that is not only out of context….
Before you make such out of context ridiculous comments read what I have written this includes my name which I have included in the bottom of my first couple of posts….
What I have said was a possible lie was your previous comment as per below.
“I might also add that I do not remember ever receiving a single email suggesting that a correctly constructed and installed DFA did not work exactly as suggested in my original articles. We have certainly received emails from people who couldn’t get it correctly bench calibrated (ie there were construction problems), but none from anyone saying that, after it was correctly set up, there were problems.”
I know of 3 other people that have had issues with the DFA and I know for a fact that at least one of the others have emailed you about the issues experienced. Of the 3 people I have personally communicated with 2 of them and there issues were exactly the same as what I experienced.
Given I know for a fact that at least 2 people have contact you with issues this statement is either a lie or an error in your memory. Given I have emailed you and John Clark I find it hard to believe you wouldn’t remember.
I have also read a comment posted on an internet forum about 2 years ago with someone having issues with there kit. Someone (in a sarcastic way) suggested they should email you for help…..
“To put it absolutely as politely as I can, the issues you apparently have had are not ones that I have shared. Clearly, you were doing something wrong, whether in the kit’s construction or in its application.”
Given your refusal to answer questions and you continue to come back with out of context statements and further questions, it is clear that I have not only been in the ball park with my previous statement but I have hit the nail on the head…. “Since you haven’t answered any of my questions you ask a question in anger I can only assume you have had much criticism on the DFA.”
I will repeat myself on this point there is nothing wrong with my kit from a construction perspective, it operates as it should and can be increased or decreased on a test bench as it should as can be done when doing real world tuning.
Even though this is the case I will be more than happy to send it to you for your personal evaluation.
The kit has been used as per the suggested application for the voltage output AFM sensor.
I used a DIY approach (I would have thought you would be happy to embrace such efforts but clearly you are just as negative to a novel approach as the type of ‘expert’ you are criticising with this column) to enable the kit to be used with a frequency output AFM.
on May 4th, 2009 at 2:04 pm
#
BG said,
on April 23rd, 2009 at 8:01 am
surely that you’re testing the wrong hypothesis or made wrong assumptions before starting the test?
Ahmm no! It’s the probability of making a type 2 error and the value is beta. IE a number between 0 and 1 or a percentage below 100%.
IE false negatives, eg. Large class sizes at schools don’t reduce student scores.
All amplifiers sound alike.
MP3’s are indistinguishable from CD’s.
Insufficient trials / participants is the cause.
Timbo
on May 29th, 2009 at 12:25 am
Julian, your lack of response has been noted.
on June 1st, 2009 at 5:30 pm
Tim,
Nice paradox… I thought you agreed with the intent of the original article.
If you knew your sample size was too small before you started the test, then surely you’ve made some bad assumptions. Or if the test was stronger, I guess this means the hypothesis was weak- then the sample size could be smaller?
So since I’m not a statistician, maybe you can explain it to us all a bit more clearly.
on June 12th, 2009 at 5:17 pm
BG
It was a general comment about hypothesis testing.
getting n / sample size up is often expensive, but it is the only way to lower the prob’s of BOTH types of error for any confidence limit we might chose.
trying to lower alpha the p of type 1 error, is the cause of high beta, because lowering alpha with a known n by setting the confidence limit tightly, will also raise beta, the p of a type 2 error. Far too often to well above 0.5.
There’s nothing wrong with 90/10% confidence limits especially if you want to avoid both type 1 and type 2 error, especially where N is lower than we’d like.
Far too many ‘scientists’ worry far too much about keeping type 1 error low, and fail to notice that they’ve made the test so ‘strong’ in your words (?), that they’ve failed to prove anything.
If Beta is forced above 0.5 then the test results are unable to ‘prove’ anything.
The validity or logic or sense of the original or null hypotheses doesn’t affect this issue. It is a product of low n and too tight a p/confidence value for the test, which does decrease the p of a type 1 error, but also increases the p of a type 2.
Isn’t nature, as revealed to us by maths, clever!?
It seems to me that both types of errors matter, and I’d prefer both values to be reported in ALL claims of proofs using hypothesis testing.
neither am I a statistician, but I did get a HD, 2D’s and 2 Cr’s at Uni in the maths/stats part of my degree. Had to reteach myself maths to do it. I was 40 when I started it.
Our values, like the truth, need to operate even in science and technology.
Objective man is a figment of people’s imaginations.
Offline? for more!? tim-bailey@actewagl.net.au
;~)!
Tim B
on September 9th, 2009 at 12:54 pm
“I’ll add something else. I think that kit was the most extensively tested electronic kit I have ever been associated with. In addition to the Prius, LS400 and Maxima, I also personally fitted and tested the kit on a Nissan 200SX, Impreza WRX and BMW 735i. ”
I can vouch for the DFA. Just had it uned on my R34 GT-T. With a bit more boost i managed to go from 161.7rwkw (Without DFA) to 177rwkw. AFR’s are consistent and safe. The car feels so much better higher in the rev range.