Read a lot about “overclocking” the ECU and changing the crystal from 4.00Mhz clock speed to 4.19Mhz clock speed…
Have seen some dyno charts done by Nick at SKUZZLE motorsport and decided to take the plunge and have the crystal changed
You can see the crystal in the pic above with the speed of 4.00 on the casing
This pic shows the 4.19 crystal installed (couldn’t get a clear pic of the info on the component)
Speaking to Nick at SKUZZLE its a good mod to do that gives you a bit more power
Daz
Do you know exactly what the crystal controls ?? Revs ?? How do you achieve more power??
Its very complicated and difficult to explain - might be better to do a google 
It leans out the AFR over 4000rpm which produces more power and raises the red line to nearly 8000rpm
Loads of good info here: http://www.mx5nutz.com/forum/index.php?showtopic=108042&page=1
Dyno graphs on page 6
Fitted the 4.19mhz ecu after a drive with the stock ecu…
You can certainly tell the difference - the car is more eager and seems to rev quicker above 4000rpm!
So a good cheap mod 
Daz
Hi
just read the thread, was the crystal ever tested on a mk2 1.8?
if so what were the results and is it available?
Andy
In depth reply -
All computers work on the same principle - clock speed. A car ECU is no different to any other micro-controller, they all work the same way.
While I’m not an auto electrician, I have worked with micro-controllers for over twenty-five years, so I know what I’m talking about. The actual program in an ECU is fixed - it’s called ROM memory. (Read Only Memory)It’s installed by the manufacturer and cannot be changed. As an example, the red line speed is fixed, as are many other factors including acceleration ramps and deceleration ramps. Air is checked at the intake (via the Intake air temp sensor), as well as the mass (air flow) these factors are compared with fuel settings and fueling is adjusted accordingly.
A micro-controller works internally with three separate components, the Input, the fixed program and the output. The ‘Clock’ is a crystal controlled device which scans each section in turn, the procedure is -
1) scan all the inputs in turn. 2)scan the fixed program, which contains all the parameters which determine what the micro-controller does. 3)Scan all the outputs and check their current status.Depending on the clock speed, this still takes several micro-seconds, and depending on the number of I/O (inputs and outputs) plus the actual fixed program, can take a mili-second or more to complete the required action. None of this is discernible, it happens faster than you can blink.Clock speed is therefore immaterial, a difference of .19 above the normal clock speed cannot be measured except by an oscilloscope, and that has to be ‘frozen’ to see it.
What you have to consider is why the manufacturer opted to use an ECU instead of the old fashioned way - that doesn’t take a genius to figure out - production engines are rarely exactly the same, they are assembled from a parts bin that may have come from different sources, possibly even individual components, though a good manufacturer will have issued strict specifications of tolerance, material and quality control to it’s suppliers. With production volumes it isn’t difficult to figure out there will be some differences from one engine to the next, and an ECU with it’s fixed parameters sorts those out, so they all perform in the same way. Any owner of an older car built from old technology (none!) will tell you there are good engines, mediocre engines and crap engines - it depends on the fall of the dice. However, most engines were acceptable. They work - was a good criteria! If you wanted a better engine - well you wen’t to an independent engine builder.
You have to consider why there are independent engine builders who supply the race industry, those engines are meticulously assembled from the finest components and put together with due care and attention - each one is tested thoroughly before delivery so it meets the end users specification. You cannot do this with production engines. Time is the ruling factor.
So - end conclusion: one word will do, however this is a family forum, so I’ll use several.
The original designer of a micro-controller has to work with different components which each have their own operating speed, computer nerds will recognize this fact from the advances in component speed and clock speed - though the latter is rarely talked about. The clock speed (crystal selection) has to match the operating speed of each component and is carefully selected. Normally increasing the clock speed by any large amount would result in a non-operating computer, unless all other components are brought up to scratch to match it. In this case the difference is marginal, but also debatable, a difference so small it’s hardly worth the effort. Faster scan time? try and measure it!
So - higher red line? - No its still the same. Better fueling? - No - fixed parameters.Clock speed cannot overcome processor settings.Nothing changes except a very slight decrease in scan time - dependant on clock speed. To believe in anything else is naive.
One word would have said it faster - and better;-)
Standard Nutz chav remediality then
Yup! Only ECU I’m aware of is the FM one, it has variable parameters that can be written in by external (PC) connection.
Debatable if it’s worth the cost, some buyers don’t know what the best settings are for any change in conditions. Once you’ve been to top dyno ‘shoot outs’ and some track events you’ll know what I mean. (Watching FM owners struggling to get the best performance on the day)
You have to start with the right engine - - - - -
Thanks for the info, very informative.
So what you’re saying is the 4.19 crystal mod is a complete placebo? how can skuzzle manage to eek a little more power by overclocking if the ROM hasn’t changed? His dyno runs and evidence suggests there is a marginal improvement. I’m not discrediting skuzzle, he does great work, I’m just failing to match what you’re saying with the results of the mod by skuzzle.
Only concession I’d make to your comment is that the sensitivity of the clock speed would be SLIGHTLY improved - it’s detecting changes in input SLIGHTLY quicker than the normal crystal. The scan time is decreased, so input changes are slightly faster detected. However, that doesn’t alter the basic fact that the program is fixed, so other than tiny adjustments (which cannot be perceived) are possible.
I’ve had three dyno checks on my car, one at one source, and two at another, when none of the readings were exactly the same. No changes at all to the car on any test. The latter (Nobles at Chesterfield) was in fact disputed by most of the owners there, we had the feeling generally they didn’t know what adjustments to make to obtain what should have been better readings for most cars.I was invited to Nobles so my car could be rated as ‘Stock’ against the rest, which were turbo and S/Cd. This was a ‘Top Dog’ contest and should have proved who had the best set up - it didn’t conclusively.
So - Dyno checks aren’t necessarily conclusive, it depends on the equipment, the operator, the weather and various factors which might seem unrelated but are.
I don’t know Skuzzle, so I cannot make any comment on the reliability of his claim, but you should ask him
when, where and how his dyno checks were done - I.E. The same place? with and without the upgrade crystal at the same time? What was the ambient temperature? Was it hot, cold, inbetween, raining or sunny? All are relevant factors.And was it the same operator (in the car and on the dyno?) Was the engine and drive train moderately warm or hot - for both crystals? Oh - and the same fuel in the tank?
Most of all - was the operator honest? You can alter settings on a dyno to suit the occasion - if a less than honest operator wanted to convince someone (who already appears convinced) that he has a better set up, then you can make it happen - the owner goes away happy, and the operator pockets the cash.
I’ll throw in a personal observation - Roll bars. Mine is a Hard Dog, one of the best, built like a Sherman tank. Owners who had fitted a roll bar before me all claimed they stiffened up the rear of the car, I convinced myself the same after I fitted mine. To be brutally honest - the only way you can check this is with strain gauges, with which I’m not familiar nor would I know how to apply them or use them, though I have a friend back in Canada who used to stick strain gauges on 727 wings which were made at the former AVRO plant in Canada. - They were thoroughly tested against strict specs from Boeing before delivery to the States. You believe what you want to believe, so placebo effect - definitely. However, main reason I fitted the bar was for safety, and after what I’ve seen, read and heard after thirteen years in this Club then I’m happy I did - better than stiffening it’s a reassurance that if anything happens I’m safe (it says here - - -!)
The ball is in Skuzzles court, prove it. Redline is as is, fueling ditto. Power increase, without engine mods - not possible. All tests must be equal, same time, same place,same weather etc etc.
How much is “marginal”?