Viewing post categorized under: Uncategorized



Feb. 25, 2017

sander, what are you doing?

Leave a comment

I will tell you, but first, a story:

For those that may not know me very well, I’m a pretty avid cyclist and general two-wheeled user of things. My interest mainly lies in mountain bikes and most things involved with dirt, but occasionally I’ve been known to do the group road ride or two. A few years ago I was sweeping (at the back) a group ride of about 8-9 people near my house and toward the end of the ride, I was plowed in to (rear ended at around 35-40 mph) by a young driver. I was knocked out, and was banged up pretty bad. Split my helmet, too. Had I not been wearing it that night, I would almost surely be in a different position / state of life at this moment.

Sidenote: bike frame was good, though (for those wondering). Steel is real, man!

Since then, I’ve been admittedly timid about getting back out on the road, but I’m slowly getting my head around being back out and have done a number of really stellar road rides since then. One of the things I’ve noticed as I’ve ridden / driven around in the years since then is that people are paying less and less attention to the act of driving. Cell phones, our ever increasing need to be busy and productive during every second of our lives, and the arguably increasing number of gadgets in cars today are preventing a genuine focus of driving.

Now, if you’re thinking, Sander, WTF are you getting at here? I’ll wrap it up.

An opportunity appeared a few months ago to start doing some research work on the vehicle integration / control side of an autonomous vehicle. Honestly, I didn’t think about that whole bike thing when I initially agreed to the research work (I mainly just thought the mental food / challenge was completely awesome), but as I’ve come to understand the challenges (chewy, incredibly interesting challenges) and numerous potential advantages to autonomy it’s made me revisit that whole scenario. I am 100% confident that that incident could have been prevented with technology that’s currently in rapid development or even already available.

The point of this is that I’ve agreed to take a full time position at the company who hired me to do the research work a few months back (I’m intentionally holding the name of the company back until a later date).

Internal combustion control will always be my first professional love, but it feels pretty stellar to work on something that could actually aid in saving someone’s life.

Obsidian Motorsport Group will still be “open” but just in a very limited capacity.

If there are any questions, please email me directly from the contact page.

Thank you to all of my customers and friends who have helped me to get to this point. Truly.

scatter_snip

Nov. 16, 2015

you should really know what you’re doing

, , Leave a comment

you should really know what you’re doing.

11/16/2015

We’re not all born with knowledge. We’ve got to work for it.

I guess I could have entitled this post “really useful stuff you can do with a simulator”, but I think that the current title is equally effective.

A long while before this current moment, most of my pre Obsidian Motorsport Group professional life was based around figuring out problems. (I promise that I’ll try and keep the retrospective of my younger self to a minimum). I started my automotive life as a sophomore high school kid working at an automotive repair shop. They did general repair and specialized in fixing old air-cooled VW and Porsche engines. Looking back on that job, I think that it really gave me a solid foundation for my professional life (and arguably my personal life, too), but at the time that I was there, I hated it.

The reason for all of my misplaced adolescent hatred was based around the idea that my bosses (two guys) wouldn’t ever allow me to just replace a part to see if the problem got better. They required me to prove that every part I replaced, needed to be replaced. If that involved showing them there was a definitive dead short in a harness with an ohm meter, or showing a oscilloscope capture of an M112/113 Mercedes crankshaft position sensor as it loses its amplitude when the sensor gets hot. Again, hated it at the time, but it really beat in to my brain that it’s best to prove your hypothesis before jumping to a conclusion.

jump_to_conclusions_mat

(…I just had a great idea)

Fast forward to now, the mindset of proving your actions with quantitative data is really the basis of what I’m trying to demonstrate the power of.

Here are some basic examples of things that are made easier with a simulator:

4344_small

Do you know how the VE model works?

Fueling is (for me, anyway) one of the more interesting variables in an engine management system to wrap ones head around. As a basic recap, fuel injector pulse width is the output of an equation inside an ECU taking in to account a number of different sensor values. IPW is the amount of time that the ECU commands the injector to be open. The amount of fuel that is delivered from the injector during this on time is based on fuel pressure, measured injector flow rate, battery voltage, and density.

The following truth is unfortunate: You don’t need to know about any of that stuff (or the ideal gas law) to tune an engine mostly well with newer volumetric efficiency based ECU’s

You can pretty much input a few simple variables and set the VE table to 90 and the engine should start up and run mostly well.
So perhaps being ignorant of the ideal gas law or what determines injector on time, figure you’ve bumbled through the map enough to get the car running pretty solid and making a “number” for the customer. Great. Good for you. Say the customer comes back and wants you to install a flex fuel sensor and re-tune for E85 or E98 or something of the sort. Are you confident that the ECU is going to add the appropriate amount of fuel over your pump gas tune when you go to make that first dyno pull on E85? What about using multiple ignition tables? Do you know that the blend table is going to work correctly? What happens if your VE table is only scaled out to 20 psiG, and you have a run out to 25 psiG, what’s the ECU going to do if you ride off the end of the table?

That can all be simulated, on your desk while you’re sitting in your favorite thinking chair.

Beyond the initial benefit of just seeing what’s going to happen to the injector pulse if you add in some ethanol content or drive off the end of the load axis, you can spend the time to reverse engineer the fuel model that is being used in the ECU. I may be a bit of an outlier in this department, but I try not to trust or have faith in anything I don’t understand or haven’t tested, first.

What about fuel compensations / trims?

Not all ECU’s are created equal in this department. How do you know if the ECU uses a multiplier or a percentage value? You would think that this would be easy to find in some ECU software programs, but it is not always as easy as one would think. It’s a lot more relaxing to do that with a simulator and move independent values around to see how it is affected before you get to the dyno / track.
Are these the same? (Emtron and MoTeC)

(picture will be replaced, soon)

I should take a second here and give an honorable mention to Life Racing’s LifeCal. They have an awesome, easy to read layout of the entire fuel calculation. See picture below:

life_fuel_calc

What about secondary injection / blend tables?

Not all ECU’s are created the same here, either. Some of the wording is a little opaque on the Mx00 MoTeC ECU’s when it comes to secondary injection

What about boost control?

If you’re the type of guy or girl who just likes to “let it eat” with open loop wastegate duty cycle controls, then this probably doesn’t matter much to you.
However, if you take pride in what you do and find a need for precise boost control, getting a feeling for the PID sensitivity controls is much more relaxing at home with a cup of coffee or any other preferred beverage of choice. PID settings are ECU manufacturer specific. Settings that work in an AEM V2 / EMS4 box WILL NOT work in an MX00 MoTeC ECU and vice-versa.

What about traction control?

Yes. This is where simulators are pretty great. The MoTeC M800 ECU’s have awesome traction control settings. The power of the M800 is the ability to make tables and compensations acting on other tables or compensations. After you get through a few levels of that, it can get a bit heavy to wrap your head around. So if you are using one of the Intermediate or Advanced simulators, you can simulate a driven wheel speed to see what’s going to happen with the timing, drive by wire throttle, fueling, or limiters BEFORE you get to the track. If you take a second to think about all of the time that can save you, it’s pretty immense.

What about setting up CAN devices to talk to each other?

You CAN do that pretty easily (I know, I’m hilarious). CAN stuff isn’t bad to do if you’re using all MoTeC products, as they provide really easy to set up templates. However it CAN get a BIT (sorry, I’ll really stop now) more difficult if you’re using MoTeC dash loggers and an AEM ECU (or Cosworth ICD and a Bosch Motorsport ECU or any number of other combinations) If you wanted to utilize all of the CAN data sent over from the AEM ECU, you would have to write a custom can template. This is not too bad to do if you can simulate values (say RPM for instance) in to the AEM ECU, connect the MoTeC dash to the AEM ECU via CAN and then work on the scalars and multipliers until the value is correct in the MoTeC dash.

So, why did I start this post by talking about myself as a young all-knowing teenager? Because I used to think that it was stupid to have to prove your work. If I could get the car fixed by guessing and maybe getting lucky, then that was good enough. The truth is that it’s really not good enough. If you can get by moving numbers around and making some power and you’re ok with that, then great. Although, I think that what’s going to separate “tuners” from professionals is knowledge and understanding. Buying a simulator isn’t going to give you the knowledge and understanding, but it can help you gain it much easier and much more safely. (and, it’s entertaining if you’re easily entertained).

2015-02-10 18.50.05

May. 15, 2015

knowledge reigns supreme over nearly everybody

, , , Leave a comment

knowledge reigns supreme, over nearly everybody

5/5/2015

I have been spending way too much time recently thinking about how I can convey the importance of data systems to you, my loyal readership. It’s the most important tool a team has in making strides in performance, diagnosis, and reliability. That’s really the best thing I can come up with. It’s used in EVERY professional race series, regardless of category. If that’s not enough to sway you toward a data system, then I’ll continue to ramble about it below.

I am in an interesting position in that I do private label work for a number of shops, teams, and individual customers. Basically, I get to see a lot of different stuff. I’ve done data analysis for a NASCAR Sprint Cup team, I’ve done analysis work for club level road racers, snowmobiles, drag racers, and a whole load of high performance street cars. The thing that all of these groups and categories have in common is that they all require some type of way to review and analyze data. Engine mapping, troubleshooting, and performance are all quantified and adjusted based around data. This all sounds super great, right? It is. Super great. But, it seems that there is a disconnect when it comes time for my customers to purchase one of these systems. Here are two problems that I’ve noticed over time.

The perceived gain problem.

There is a big problem with the understanding behind purchasing data systems. The problem, as far as I see it, is that there’s no direct connection in a customer’s mind behind purchasing a data package and performance gained.

For instance, customer A wants to buy a turbocharger that has a 50lb/min max flow rating. His current turbo is 30lb/min. This will yield an approximate 200 hp gain (all things being fair and equal in the world of physics). Customer A will go to a dyno, get his engine mapped, and there will be a before and after chart representing the difference in power. He will see and feel the money that he spent. With 200 extra hp, maybe he’ll drop 2-3 seconds off his lap time. Maybe. That’s if the extra power doesn’t through the cars balance off, or over drive the tire, or over heat the brakes.

What happens if the customer is told that he can spend the same amount of money on a data system that could take 5.00 seconds off his lap time by using accelerometer data and driving line evaluation, potentially identify faulty sensors, potentially identifying an engine failure BEFORE it happens, and allow for peace of mind if all of the data looks great?

He’s going to go for the turbocharger.

The problem at hand is that there is no perceived direct correlation between data and performance. It’s unfortunate because, when used correctly, this couldn’t be further from the truth.

Now it’s probably obvious by reading the other content on my website that I sell data systems, and it would be great if you contacted me to talk about them in more detail. But, all bias aside, they really are useful pieces of equipment that sometimes can result in the dramatic lap time reduction mentioned above, among other useful purposes.

Thoughts on being emotionless.

There’s such a nice feeling associated with making decisions based off of objective data. It completely removes the need for “emotion” or “feelings”. In the context of engine mapping, this concept couldn’t be more true and to the point. Imagine you’re sitting in the drivers seat of a 1000 hp Supra strapped to the dyno and you’ve just started a pull in 4th gear. You’ve got a lot going through your mind. “Does this engine have a coolant leak?” “Does it actually have E85 in it, or is it diesel?” “What’s that sound? Rod knock? Valve tap? Oil spraying out of the cylinder head?” The list goes on and on. And then you’ve got the normal stuff you’re supposed to look at like lambda, fuel pressure, temps, etc…

How to remove some of those points of concern:

-If you have cooling system pressure sensors, you can derive if theres a leak, or if you have a head gasket that’s starting to leak, or leaking profusely.

-If you have an ethanol content analyzer, you can tell exact ethanol content and not have to worry if you’ve got winter blend E60 or E70 as opposed to the E85 that you should have.

-Oil pressure sensors in different points of the engine can tell you a lot about the oil system efficiency or if there’s any oil in the engine at all (…)

The more you can do to remove these questions, the more mental focus you can apply to doing the original job of calibrating the engine. Furthermore, you can set up safe guards with good ECU’s that can impose a fuel cut, a warning light, or a boost reduction if any number of things are out of your specified range.

See below for an example. In this instance the AEM V2 ECU’s allow you to create a predicted path for fuel and oil pressure. It’s pretty simple but cleanly executed. Oil pressure changes with engine speed and thus, you can create a low oil pressure threshold at different RPM’s to compensate for that. I.E. if the oil pressure drops below x oil pressure at y rpm, the engine will cut for a brief period of time. Same goes with fuel pressure. At different manifold pressures, 1:1 fuel pressure regulators allow for 1 pound more fuel pressure for each 1 pound of manifold pressure (a calculated differential fuel pressure channel would remove this requirement entirely, but that’s maybe for another article, some day). As manifold pressure rises, so does the fuel pressure low threshold.

fuel_press_protect
Just realized that I’m rambling a little bit, sorry.

So, I’ve explained how extra sensors or a data analysis package could help someone like me (who is someone who maps engines, regularly). Unless you’re like me, this probably doesn’t interest you tremendously.

A thing that is tremendously useful for road racer types is the time difference graph. Have you ever wondered if you’re making an improvement on track? Or even better, exactly where? Practically all worthwhile GPS based data systems give you this information. Attached below is a screen shot of two laps overlayed on each other.

sled_compare

The upper graph is a representation of transmission speed, and engine speed, from two separate laps. The lower graph is the time difference graph. The blue line represents the faster lap time and it is used as the reference. The red line is the slower lap time. Since these two are overlayed based on distance, they are equally located based on position. This allows for an easy representation of EXACTLY WHERE you’re gaining or losing time. If the red value is positive the vehicle is losing time over the faster lap, if the red value is negative, the vehicle is gaining time (going quicker) over the faster lap.

Even cooler than this, is that the software will auto generate a track map based off of GPS position. And since all of the data is located based on GPS coordinates you can select anywhere in the data and it will show up exactly where you’re at on the track at that particular data point.

So for example, the black circled area signifies where the cursor is selecting data.

sled_compare

Now all you need to do is to transfer over to the GPS track map and you will have an exact identification of where on track the selected data is correlated to. Neat.

track_pos

Now take a few minutes to think about the usefulness of all of this information…
GPS based data systems can start around $1200.00. Fully featured GPS based dash loggers can start around $2500.00. (Be sure to ask me about all of the other functions that the dash loggers that I sell (from MoTeC) can do.

Pretty affordable if you really think about it.
Please feel free to email me any questions you might have. I’d love to chat.

Jan. 30, 2015

mcmuffins and half bridges

Leave a comment

mcmuffins and half bridges

1/30/2015

Things you can do with a MoTeC Dual Half Bridge:

41213

The MoTeC dual half bridge is a pretty neat device. Simply put, it allows you to drive reasonably large, variable speed motors, directly from the ECU. Why do you care about this? I’ll tell you why, because the world isn’t as simple as “ON” or “OFF”.

The DHB is effectively a solid state relay that can take a low current PWM (pulse width modulated) signal (driven low or high) from an ECU or a Dash and drive a very high current (20A continuous, maximum) motor at varying speeds or positions. There are a number of different cool circumstances where you could use a DHB to your advantage. I’ll run down a few below:

Example 1:

We’ve all heard that car. You know the car I’m talking about. The car with the big ass radiator fan that sounds like a jet is about to take off. I find that so annoying. No one needs a loud ass fan to turn on for 60 seconds just to cool your radiator off, so that it can turn back on again, 2 minutes later.

For instance, when you go to order your favorite McBreakfast from the McResturant, you don’t pull up to the window and scream at the top of your lungs “I WANT A MCMUFFIN”. You say, “I want a mcmuffin, please”. Your radiator fan is screaming “I WANT A MCMUFFIN” every time it turns on full blast for no particular emergency. Drives me nuts.
So the DHB can fix this problem. The DHB can do the high current work of fan speed control. All ECU’s that are worthwhile will allow the user to create a PWM output table based on user definable axis. So then just create a table that looks like this:

water_pump_speed

You’ll have coolant temperature on the X-axis and the values in the table will correspond to fan duty cycle (or speed). As temperature increases, so will fan speed until it gets to the point of being at full blast. The thing to note, is that it will only be turning at full speed (and screaming about the MCMUFFIN) if it NEEDS to be there. Cool.

Example 2:

Say you’re an R35 GTR owner, or a Supra owner, or a huge power engine owner, and you have 3 fuel pumps to supply your daily driven 1000+ hp engine. If you have 3 fuel pumps running at all times, you’re wasting energy in two ways.

-First, you’re going to waste a ton of electrical energy. Figure worst case scenario your pumps draw 20A each.

-3 fuel pumps. 20A a piece (average worst case scenario for your average 340-400LPH pump)

-20A times 3 pumps = 60A. Figure 13 volts while the engine is running. Amps times volts equals watts. So, 60 * 13 = 780 watts of power to drive your three fuel pumps at idle. If you’re like most in the US, watts don’t really mean much to you in an automotive world.

(sidenote of relevance) the Kilowatt is used to measure Horsepower in other parts of the world.

-However, for this example, it is relevant.
1 watt = 0.00134 horsepowers

So after doing some simple math, we can figure that those three fuel pumps are taking a total of ~1 horsepower to drive. That’s crazy. You don’t need that when you’re idling!

-Secondly, you’re going to heat up the fuel excessively. Fuel pressure regulators are just like any other regulating valve, they restrict the flow so that pressure can increase, and they allow less restriction so that pressure can decrease. The variability is based off of manifold pressure / vacuum (that’s what that hose from your intake manifold to your fuel pressure regulator does!).

So now think about your fuel pumps trying to force the maximum volume of fuel that they can flow through a tiny regulator and fuel injectors that are barely opening (because you’re at idle). This is a strain on the pumps, and when there is strain, there is heat. This heat will eventually be transferred in to your fuel system and you will watch your fuel temperatures steadily rise. Hot fuel isn’t so good, but not totally deadly. What can be bad is that if your car was tuned on a dyno with cooler fuel temperatures, and then you go out and drive your car around the city for an hour and then have a blast on a local highway, you might find that your mixtures are slightly leaner due to fuel density changes relative to heat. (Injector dynamics wrote a really great article about this, it’s short, and to the point injectordynamics.com/articles/injector-dynamics-newsletter-feb-2013/ )

Ok, now we know that running your fuel pumps at full bore all the time maybe isn’t such a great idea. Now what do we do?
What if you could slow down your fuel pumps at idle and part throttle conditions? What if you could have them speed up as you creep in to boost?

You can!

All ECU’s that are worthwhile will allow the user to create a PWM output table based on user definable axis. So then just create a table that looks like this:

fuel_pump_speed

(Sidenote of relevance: Even this table is over complicated. You can just have this referenced to manifold pressure without TPS, I just think it’s neat to be able to change the speed up relative to TPS too)

The DHB can take this small current control output from the ECU and drive the heavy current fuel pump with it. So when you’re on the throttle 100% at 0 manifold vacuum, the fuel pump duty cycle will be 61%, and will gradually increase to 100% by the time you reach 175kpa (a little less than 30 psi of boost).

Use your parts efficiently!

Shameless plug is that I usually have one of these around for sale, this moment in time is no exception. Email me if you’re interested in hearing more about it, or just want more of my rants. sander@obsidianeng.com

Nov. 10, 2014

it’s about time

, Leave a comment

it’s about time.

11/10/2014

The concept of time is something that people have written about since the beginning of time (sorry, I know that’s awkward to read). Time with regard to engine management systems and fuel injection isn’t something people really give too much thought to, though. I thought it would be nice to break down how important time is with fuel injection and engine management (we’ll just mainly focus on fuel injection time in this article.)

Combustion events happen pretty quickly, even with arguably boring engines, like the one in your mom’s mini-van. The point and purpose of this article is to show you how fast these event happen and why that matters to you and your injector choice.

The basics of port fuel injectors are pretty simple:

– (part 1) You have an injector that will flow a given amount of volume of fuel per minute. (Most know this is lb/hour, or cc/min).

– (part 2) The injector can flow during the entire two crankshaft revolution engine cycle. (Remember that a 4-stroke engine requires two complete revolutions to complete an engine cycle). If the injector is open during this entire time, this is represented as 100% Injector Duty Cycle (or 100% IDC).

100% IDC should be avoided at all costs. Bad things can happen to the injector at IDC above 95%. A safe place to
wind up is around 85% max.

– (part 3) This engine cycle time calculation is really easy to do.

Part 1:
If you buy a set of injectors from whom ever (but let’s be honest, you really should buy injectors from ID www.injectordynamics.com *end of shameless plug*), they will come with a flow rating, 1000cc/min, 500lb/hour, etc. What that means is that a 1000cc / min injector can flow 1000 cubic centimeters of fuel in one minute. The neat thing about that is that pretty much all injectors have a linear flow relationship above 2 milliseconds of ON time. Let’s take an example:

You have an injector that flows 60 pounds of fuel per hour (630cc per minute for you metric folks). If you divide 60 pounds of fuel per hour by 60 minutes you will get 1 pound of fuel per minute.

If you know that you need 0.6 lbs of fuel per minute to achieve a certain air fuel ratio, at a certain rpm (say 8000). you can easily derive from that that you will have to operate at 60% duty cycle to flow 0.6 lbs of fuel per minute.

Sidenote: There are really sweet equations to determine air flow based off of some standards and some values like displacement, VE, fuel type, and some others. If there are enough people interested I’ll walk through the entire process. It’s neat but it’s lengthy! You can actually calculate that required fuel per minute that is listed above (0.6 lbs of fuel per minute).

Part 2:
A 4 stroke engine cycle is exactly that, 4 individual events that must occur in sequential order to complete one engine cycle.

Intake stroke (as the piston travels down)
Compression stroke (as the piston travels back up)
Power stroke (as the piston travels back down after the explosion)
Exhaust stroke (as the piston travels back up to expel the exhaust gases).

On a port fuel injection car, the fuel injector is behind the intake valve and therefore, the fuel allowed to enter the engine is really ultimately controlled by the intake valve position. The fuel injector can stay open for the entire engine cycle (all 4 strokes) and fuel will only travel in to the cylinder to be ignited when the intake valve is open. If it’s on for the entire engine cycle, this represents 100% injector duty cycle.

Sidenote: Direct injection is now in a number of cars and this principal is completely different, as the injector nozzle is DIRECTLY (see what I did there?) in the combustion chamber. The maximum duty cycle that they can operate at is around 50%. Direct injection is pretty neat to think about considering the duty cycle limitations, the 400-3000 psi fuel pressure ranges, and the fact that the OE ECU’s are controlling multiple injections per cycle depending on load and engine speed, wild.

Part 3:
So, we learned in part 1 that we have an imaginary engine that needs 0.6 lbs of fuel per minute to achieve a certain air fuel ratio (or AFR) at 8000 rpm, and that this will require 60% injector duty cycle to do so based on using a 60 lb per minute injector.

That doesn’t really even mean a whole lot unless you can tell the fuel injector how long to stay on for. 60% IDC doesn’t really mean a whole lot to an ECU. It does in some cases, but not really with injectors. So, let’s figure out how much time exactly 60% IDC means, or more specifically, how much time exactly does the injector need to be open for.

Lets say our engine is at 8000 RPM. We want to figure out how much time one engine cycle takes (4 strokes).

We first need to take the engine speed and divide it by 2 to get the number cycles. The 2 is there because it takes 2 revolutions to make one engine cycle.

8000 / 2 = 4000 engine cycles per minute.

Then we want to figure out the number of cycles per second. So we’ll divide the cycles per minute by 60 seconds to find out the cycles per second.

4000 / 60 = 66.66 cycles per second. (pretty wild if you think about it, right?)

Then we want to figure out the time it takes for each engine cycle. So we can just divide 1 by the number of cycles per second.

1 / 66.66 = 0.015 seconds per cycle. This equates to 15 milliseconds per cycle.

So now we know how long it takes for each cycle at 8000 rpm. So we know from above that our imaginary engine needs 0.6 lb/min of fuel to obtain our target AFR at 8000 rpm, and we know that it will require 60% duty cycle to flow 0.6 lb/min with a 60 lb/hour fuel injector.

So now it’s easy! We just need to multiply the time that the engine cycle takes by the duty cycle (divided by 100)

(15 ms * (60/100)) = 9 milliseconds

Are you still reading? I hope you’re not more confused than when you started.

So I don’t get accused (again) of creating a boring looking wall of text, here’s a graph showing engine cycle time vs. RPM.

cycle_time

Here’s a fun fact to end it with. Moto GP bikes have rev limiters around 17,500-18,000 RPM. If you calculate it out, at 18,000 rpm, that engine completes two revolutions in 6.6 milliseconds. That’s 0.006 seconds! Whoa.

Jun. 12, 2014

products and an ethos

, , Leave a comment

products and an ethos.

6/12/2014

There are a small number of things in this world that really get me excited.
Here’s a short list:

-Fuel injection systems, and internal combustion. (well, obviously)
-Race car electronics.
-Bicycles.
-Design work.

I love all of these things equally. Honestly, they’re all just so fascinating to think about. There have been some situations in the past where I’ve been able to combine two of these things. For instance, I designed and produced a suspension linkage for a bicycle I used to own. The suspension design on that bike was pretty good, but it was far (very far) from perfect. The bike had a very poorly machined rocker assembly that made for a loose connection to the rear triangle. A friend of mine said it would be way better if they had used a sealed bearing instead of an IGUS bushing (a thin plastic bushing).

So, off I went.

Being one of my first design projects, it took a long time. A really long time. I pushed through it and I learned more than I probably ever thought I would have. After I finished it, I submitted the design to a local machine shop and the finished pieces arrived at my house. I ripped open the box, pressed in the bearings I had spec’d for the part and it slid right in to place. I was ecstatic. That project allowed me to combine two things on that list. I loved that.

 

banshee-rc4-and-rockers_6332845523_o

Skip ahead a number of years later…

I bought a 2006 Ducati Monster S2R. I bought the bike a bit on the cheaper side due to an “intermittent stalling” problem that the previous owner had no idea what the problem was. After some time the stalling problem happened to me while I was riding it home one day. Some people may have been distraught by this, I was actually thrilled. This meant that I could finally identify and fix the problem.
After some quick electrical testing, I found that the wiring to power the fuel pump had a poor connection in the bulkhead that passes the wiring in to the tank. After some internet searching I couldn’t believe that no one made (what I felt was) a good repair for this!

So, off I went, again.

With the help of a good friend at Orlov Design, we digitized and reverse engineered the factory fuel pump mount / bulkhead assembly. I then designed a piece to be CNC machined out of a block of aluminum to replicate the factory unit and add a hermetically sealed military grade bulkhead connector. This connector is now easily serviceable if anything were to happen to the seal or the connections and it provides for an easy removal of the fuel tank with a quick turn of the electrical connector body. Also is the option to add AN thread fuel fittings if an individual wished.

 

 

After a few minor revisions, the prototype is complete and installed in my bike currently. I’ve been riding it around an have experienced no problems to speak of. After some finalizing and time spent gauging interest, the product may be offered to the public.

This most recent project allowed me to combine all of these things on my list of things in this world that get me really excited. It’s pretty easy to extrapolate how excellent this makes me feel.

The basic ethos of this company is pretty simple. Do everything the best that you can, and be excited about what you do.

If there are any questions about this, please email me from the Contact page.

Aug. 30, 2013

MPG or GPM?

, , Leave a comment

MPG? pfft.

8/30/2013

Something occurred to me the other day when I was looking through old data to answer a question that a customer had about fuel economy. The question was (basically) “if you know your ultimate post-combustion air fuel ratio, and you know the amount of air volume going in, could you calculate your fuel level in the tank with out a fuel level sensor?”

as a sidenote, I love these kinds of questions

The answer is most certainly, yes. You would need to have a way to tell the ECU when the fuel tank was full and with how many gallons (via a set parameter in the ECU), but you would be able to calculate the fuel used, fuel trims, and a bunch of other nerdy data.

This got me thinking about some of the higher horsepower cars that I’ve tuned recently. At maximum power output, a GT-R that I worked with recently consumes an almost inconceivable amount of fuel. I thought it would be cool to (quickly) show you how to calculate the rough fuel consumption by just reading some fuel injector duty cycle charts, and knowing some basic info.

This file has been lost during the great website crash of 2016. Sorry

There are a few key things that we need to identify here. All of these graphs are lined up. So, that means that all of the data is on the same time register. So for instance, if you look to the “Engine Speed [rpm]” graph, you’ll see that the value at the yellow line intersecting it is 6928.9 RPM. This 6928.9 RPM point also corresponds with the 78.635% “Fuel Injector Duty Cycle [%]”. For the purposes of this demonstration, this is basically all we need to know. However, it is cool to see the calculated torque value there at the bottom, among all of the other data points of interest.

Sidenote of relevance if you notice something odd about the fuel injector duty cycle graph. The IDC plot is only sampling at 1hz (one sample every second). I didn’t think that I would be writing this at the time when I was tuning the car, and as it turned out, the IDC sample rate was very low. For the purposes of this we will just assume that the IDC is somewhere around 80-90 percent.

So, now we know that the Fuel Injector Duty cycle is around 80% at peak power / rpm. This number alone tells us nothing. We need to know the amount of volume that the injector will pass at 80% duty cycle. Luckily for us, these are awesome injectors, from an awesome company, that publishes plenty of info about them.

id2000_data

These injectors are classified as 2000cc / minute, nominal flow. For those of you not familiar with the metric system, that means that at 100% duty cycle, these are advertised as being able to pass 2 liters of fluid in one minute. Pretty significant.

If you take a look at the chart, you’ll see a number of different fuel pressure values on the left hand side. This is representative of the different fuel pressures that these injectors can operate with, and the respective change it has to it’s ultimate flow characteristics (see the corresponding flow values on the right hand side). The GT-R, with these injectors, had a base pressure of approximately 55 psi. This 55 psi value corresponds to 2470cc/min of fuel flow at maximum duty cycle.

Yet another sidenote of importance. You’ll notice that the fuel pressure differential value listed in that screen shot is actually more like 51 psi. For the purposes of this example, we’ll use the 55 psi value for ease of calculation

 

Cool, right?

 

So we know that 100% injector duty cycle, at 55 psi of fuel pressure, equates to 2470cc/min of fuel flow. We’re at 80% injector duty cycle according to our graph (above). Now for some easy math.

2470cc/min * 0.8 (80% duty cycle) = 1976cc / min

We almost forgot the really fascinating part. this 1.97 Liters of fuel per minute at max power is only for ONE CYLINDER. We need to now multiply this by 6.

1976cc / min * 6 (6 cylinder engine, 1 injector for each cylinder) = 11.85 Liters of fuel every minute.

There are 3.785 liters in every gallon.

11.85 Liters of fuel consumed every minute / 3.785 liters in every gallon = 3.132 gallons consumed every minute

We can go one step further, and figure out how long it would take the car to run out of fuel on a full tank. After a quick Google search, I found that the GT-R fuel tank is a scant 19.5 gallons.

19.5 gallon fuel tank / 3.132 gallons consumed every minute = 6.22 minutes till empty

So, it would take this lucky guy a little less than 6.5 minutes to clean out his fuel tank at 7000 rpm and 40 psi of boost, with E85 fuel. If he had the set to stay in the gas for 5 minutes straight, I would give him a high five.

Aug. 15, 2013

it’s all about the delta

, Leave a comment

It’s all about the what?

8/15/2013

Why?

What is the delta?

By definition, the delta is the change in any changeable quantity. The delta is what sums up the basis of OEG. It is the quantified measurement of improvement over any number of given points. When you get your engine calibrated / tuned by me, you are getting the greatest delta that I can safely provide over all given rpm points, not just the peak values.

It’s unfortunate that the majority of people have an understanding that a dyno session will only reveal two numbers, maximum horsepower and maximum torque output. My understanding of this is founded from the question that I often recieve after tuning a clients car:

“What did it make?”

Now, it should be understood that the maximum power output of an engine is an important piece of data. It can be one of the class selection variables in NASA and SCCA road racing series’. It can put in perspective the overall health of the engine, but it is really only a small fraction of what really matters after a dyno tuning / calibrating session.

Figure 1 is a dyno plot of two similarly modified Nissan GT-R’s.

It should be noted that these two plots are different cars with different sized turbo chargers, and only the red plot has a properly tuned closed loop boost control algorithm.

The point of this article is only to highlight the little amount that “peak numbers” really tell you.

Figure 1 Figure 1

You’ll notice that the red horsepower and torque plots make a little less power and torque above 6750 RPM, but practically everywhere above 3700 the red plot makes substantially more horsepower and torque. What this means in the end is that the power and torque output that is felt by the driver will be much easier to access being that it makes the greater power and torque numbers at a LOWER rpm.

If you were to focus on the peak numbers alone, you might think that just because the blue plot shows that the engine made 1266 whp it would marginally faster than the engine that only (yes, “only”) made 1084 whp. I would be willing to bet given the same circumstances, that if these two cars were to drag race from a stop that the race would be very close.

Note, this is just conjecture and the race would depend greatly on gear ratios and shift points.

Now, what isn’t conjecture is that the red plot will yield a much more enjoyable car to drive. Take a look at the 5000 RPM x-axis column. The red plot is 200 ft/lbs of torque MORE than the blue plot!

The purpose of this is to show that there is so much more to calibrating / tuning an engine than just peak output numbers. I always focus on giving the customer the best outcome at all measurable points. Any questions, concerns, or comments? Please write me from the Contact page, It would be great to hear from you.