Penny-pinching car geek's guide to racing, track days, and car build. DIY projects, product reviews, and interviews.

Thursday, November 25, 2010

Life, November 2010

Left to right: Yours truly, and classmates Kevin and Nguyen


On a side note, I got married recently.  I still have the family ceremonies to do, and that is planned for after I pass the bar exam.  But as of this week, I am a married man!  Wow.  I did not see that coming.  No one else did either.  I'm getting the full law school experience; J.D., wife, what else?  I can't wait until I see what life has in store for me. 

Last weekend, I went to a bar association conference.  I also had a job interview.  Good results!  I found someone to mentor me, and I met a lot of great people.  But most importantly, I will be in California this winter working a temp job at a law firm.  I am going to do my best and see if that will take me elsewhere!  I know this sounds cheap, but seeing Ferraris and McLarens as I drive to my interview really drives me to work harder.  I don't plan on living in vanity, and I'll probably never own a Ferrari to drive around, but I do plan on having enough money to raise my family and go wheel to wheel racing.

J.W. Mariott

Wish me luck!

Tuning the RX-7 AFM - Resistor Trick

A bit of searching on the forums revealed an old but simple and effective to tune the RX-7 AFM. The RX-7 AFM runs too lean, even though it still performs better than the stock AFM. This is especially true for the lower RPM's. It would help to adjust add more fuel at all RPMs. This isn't optimum but a good compromise between simplicity and price. Thanks DVCN for the information. Basically, you will be soldering in a resistor which would adjust the AFM throughout the entire RPM range.

Stuff needed:

1. RX-7 AFM

2. 25ohm resistor (Actually, anything between 20-25ohms) - Less than a few bucks at Radioshack
3. Soldering equipment
4. Sealant (RTV works great)

Air to fuel ratio

I'll do a short write up about what Air-to-fuel ratios are. There is a certain optimum Air-to-fuel ratio in each engine. Theoretically, the most optimum ratio is called "Stoichiometric mixture," which is approximated as 14.7 parts air and 1 part fuel, with numbers representing mass.
The AFR is generally expressed in this manner. When someone says "AFR at 15" it means 15 parts air, 1 part fuel. Maximum torque usually occurs a bit under Stoic mixture. According to my co-driver Brian, who is also a mechanical engineer, this is due to the fact that air and fuel does not optimally mix. That is partly why electronic fuel injection is better than carburation, and direct injection is better than regular injection, and why all these different injection methods have been developed. Some of these injection methods attempt to create a condition where maximum torque occurs closer to the Stoic mixture. This makes the car more fuel efficient and more powerful.

Generally, a leaner burn is good for fuel economy because less energy is lost to heat. Slightly richer mixes result in more torque and cooler burn, preventing knock.
Air-fuel mixtures are usually measured with an O2 sensor. A wideband O2 sensor is crucial for this kind of tuning, not that you need one to do what I did. Anyways, a 16+ year old miata engine is far from having any sort of fancy injection technology and optimum torque occurs a bit below 13:1.

People adjust this a little bit by adjusting the spring tension of the flapper door in the AFM. Looser spring means more fuel, as the door opens more and the AFM sends a signal to the computer that more air is coming in. Of course, in reality the same amount of air comes in regardless of the flapper door angle, but it does mean that the door opens up a bit more. Usually people tune by adjusting the spring 5-7 clicks looser. I don't know how much AFR is adjusted this way, but I do know that it is less optimum than this trick. Furthermore, the spring method messes with the idle, and you have to take further steps to make the idle behave.

The Resistor Trick

DVCN, who came up with this method of tuning. He found out that adding a resistor to the circuit on the AFM adjusts the mixture. He measured the following on the wideband:


20 ohms:
15's below 2k, 13's in the midrange going to 12.7 at 7000rpms.
25 ohms:
Just slightly richer everywhere, going to 12.3:1 at 7000 rpms.
100 ohms: In the 11's at 7k.

I chose to use a 22.5ohm resistor. If you have 50 ohm resistors lyring around, you can run it parallel and you'll get 25 ohms. DVCN stated that 25ohms felt the best. I picked 22.5 because I just wanted the best top end I can get and I thought something richer than 12.7 but leaner than 12.3 would be more ideal for top end. I may be completely wrong on this.

Instructions
Anyways, here are the instructions. I did this with the AFM removed, with good lighting.


First: Open up the black cover on the RX-7 AFM. I used a screwdriver to pry out the cover. Try to be careful not to nick the plastic too much.
Second: Clear the way. The connection you want to work with is the third one from the right, with the AFM's inlet to the right of you. Bend the fourth one out of the way a bit so you can get the soldering equipment in there.

Third: Bend the resistor to the shape you think you will be using. Cut the connection appropriately. Clean the area if there are any visible impurities.

Fourth: Carefully solder the area. You don't need excess lead on the contacts. Make sure it doesn't overflow and touch other contacts.

Fifth:
Seal the cover. I gobbed on blue RTV onto the cover edges and then placed it back. I gobbed on some more RTV around the edges afterwards. It will dry to a nice rubbery seal.
Finished! Do a test run and see if you like it. I did not have to adjust idle at all; it sits okay at 850-1200 rpms. Funky because my car has power steering and AC removed, yours could be more stable.

Results? Added 1+mph on my maximum speed on my home track. Note that the second fastest speed I recorded were done on cooler temperatures. Between my mods, I accelerated faster than a 1.8l miata with a bone-stock engine.

Thursday, November 4, 2010

Sway bar short endlinks

Total Price: 19 dollars.
End link prices are pretty high. I don't really understand why. I guess when the endlink uses exotic-ish lining on the heim joints to reduce noise might get expensive. Some people build from parts they got from McMaster-Carr. Everything I find on McMaster-Carr that I found elsewhere, M-C had the higher price. Of course, McMaster-Carr does sell a huge variety of products in a single catalog that makes it easy for you to browse.  Furthermore, their customer service is excellent, and shipping is fast.  So if you need high-end rod ends quickly, say for your formula car's pushrod end,  it would make sense to go with them.  Otherwise, keep searching! Or just go with what I found.

I bought short endlinks from ISC racing before. Good price. I broke that endlink when I crashed into a curb over the summer, so I went on hunting for a replacement one.

QA-1 Racing, the guys who make the springs I have on my car, are also rod end manufacturers. Total cost is 19 dollars for two short endlinks. Check out their website. For the Miata, you want the 10mm right hand thread MCF rod ends, male and female. Two of each. You can get two jam nuts from a local hardware store for about 50 cents. You will have to source the spacers elsewhere though. I just found some spare nuts and threaded the rod end mounting bolts through it. Sure beats spending 70 dollars on a set. 

QA-1 has a minimum order of 25 dollars. But if you get both front and back sets it should clear the minimum order. They also have prompt customer service. When I asked for part weights and other specifications, I got a reply the next day with the information I wanted.

My car is really noisy to begin with, so I might not be noticing it, but I didn't notice any additional noise. Tolerances are pretty acceptable to me, measured totally scientifically by taking the rod ends and bolts and moving them around by hand. They also survived use at the local road course, and both me and my co-driver went off course that day.

Link to company: QA-1 Motorsports

Wednesday, November 3, 2010

Racechrono - Reasonably priced datalogger (GPS Lap Timer and Data Logger )

I have been timing my laps and analyzing my driving and car tuning through a GPS data logger. There are so many options out there, ranging from an enthusiast level logger such as the Performance box, to full-fledged expandable professional solutions like the Race-Technology DL-1. Prices range from 400 dollars to many thousands of dollars. Some of these products are very solid and capable. Others are kind of crap. For example, MaxQdata sells a product called "TraQr." It is priced at nearly 300 dollars. As far as I know, the unit is nothing more than an off the shelf GPS logger that is available for much less. In fact, if I recall correctly, the MaxQ unit even looks similar to an off the shelf unit with a different logo on it. And there are those units out there where you can pay a hundred dollars or more than this solution I am about to propose, for simplicity (usually these are standalone units) and a bit of added capability.

My advice is this. Do some research before you plop down cash on any data logger. A lot of research. I wouldn't have found out about this solution if I hadn't done so.

I have been running a 5hz data logger for about a year now. Recently, I switched over to a 10hz logger. Some of the things I can do with this rig are as follows: lap timing, lateral and longitutional acceleration (G-forces), lap comparison with a lap from any session, traveled route. If I spent a bit more money and had an OBD-II car, I can do even more.



So what is the cost of all this? As low as 80 dollars to a maximum of 120 dollars. Yes and it is reliable. Yes it is light weight. Yes, it can be a live timer. Yes, there is a lot of community support. All possible thanks to these guys:


(Image is a link)

Racechrono is a very capable, free, data analysis and lap timer software. It is being constantly updated and best of all, it is free! It supports Windows Mobile and Symbian operating systems on phones. Of course, there is also version for your personal computer. I won't go too much into details, but I have some screen shots on the bottom. You can also read about it on the RaceChrono website.

Items needed
So what do you need for live timing and data logging? There are several different ways to do this, but this is probably the most cost effective way.

1. A Smartphone or a PDA
Samsung Blackjack. Old, but gets the job done.
You need a smartphone that either runs on Windows Mobile 5.0+ or Nokia S60 2d ed.+ with bluetooth support. This part can be really cheap. My first phone for this rig was an old Samsung Blackjack. It cost me 50 dollars for a good condition used phone with several chargers, car chargers, extra battery, and audio output adaptor. Slim, light, got the job done. But If you are going to run the 10hz GPS receiver, I recommend a phone that runs on Symbian. This is because the 10hz receiver seems to have some issues logging at 10hz unless you set the baudrate manually through the Racechrono software, and this trick doesn't work on Windows Mobile phones. I currently run a Nokia 5340. It feels like it works much better with RaceChrono. Having a large, touch screen makes it a lot easier to use in the cockpit. I also recommend a phone with SD card support. I hate using cables to upload data. I used this website to look up phone specifications: Phonescoop

2. GPS Receiver
There are many options out there. I recommend the QSTARZ BT-Q818X for 5hz, and QSTARZ BT-Q818XT for 10hz. This is because these units are cheap and hassle free. Other units need to be set to 5hz through special software on your computer. These units have an external switch that selects the rate (Off-1hz-10hz). The price differences between the X and XT model is very small; just go ahead and get the 10hz model. If not, you have an option of getting other bluetooth receiver-only models (like the ones I mentioned) or getting one that internally logs. There is no point on getting the latter unless you want to manually export the NMEA data and upload it to your device later.

Qstarz BT-Q818XT
 Note: when you shop for the receivers, do some research to find out if the update rates are actually what they claim to be. Often times they are 3-4hz units that interpolates internally to 5hz, or it updates at a certain rate but its ability to log it doesn't match the update rate. There may be other issues as well. The units I recommended work just fine.

3. Securing the receiver
I used to use blue painter's tape. I got tired of that so I went out and bought heavy duty velcro from wal-mart. The adhesive is can be very strong, and so is the velcro. One patch at about 22mm diameter is plenty enough to hold it in place. I used two. Minor problem is that the rubber backing on the BT-Q818XT doesn't work with the adhesive. I simply peeled the backing off using my finger nail. Its tough but it will come off. 



Setup
1. Download the Racechrono file from the website
2. Download it to your phone. I used my SD card to do this.
3. Run the file using your phone's file explorer application.
4. In the RaceChrono menu, go to Settings -> GPS Receiver -> Baudrate and set the baudrate to 115200.

How to record a track
1. Start a new session. Write your session title, Driver, and then go into the Track option.
2. Select on "Create new" and write a track name.
3. Select "start" on the bottom left.
4. Drive a full lap. Don't forget to do a outlap, warm up, and a cool-down lap!

How to set start/finish line
There are two ways to do this. In-car, and after your laps. Don't do the in-car method. It is stupid and you will get off your line as you lose concentration. After you do your laps, park your car at the pits. Make sure your session is paused: if not, click on "pause" at the bottom. Browse traps -> add new -> start/finish line. Either use the directional buttons or if you have a touch screen, drag the screen left and right. You can zoom in for more accuracy. Hit "OK" at the place you want. If you are doing an autocross walk, you can set start and finish seperately as well.

How to set splits
Follow the same procedure above, except if you already have a track plotted out, you don't have to have the timer running (resume session). Add splits as desired.

Live timing
Very simple, as you start the session just hit "start" or "resume" and then hit "switch to live timer." If your phone has an accelerometer, it will recognize the position of the phone by itself and orient the display landscape or portrait. If not, you can go into the menu and set it up manually.

Overall impression
I really like it. My old 5hz unit would (probably through interpolation) match the Performance box times when I ran them together. The 10hz unit must be even more accurate. The 10hz logger is apparently an actual native 10hz that really logs at that rate. Amazing for the price! The Racechrono software is very capable and very complete; it is just as good as high-end systems are. Furthermore, you can expand it with a bluetooth OBD2 reader and log even more parameters. You can find out how to at the Racechrono website. Racechrono for your PC can import and export data to many different formats, including format that is compatible with Racelogic and other data analysis software. This adds so much versatility to this system. And it is so affordable. If you are a professional team that needs to log shock position at 100hz, have steering and brake position sensors, and need 20hz+ GPS log rate, this unit is obviously not for you. But if you are an enthusiast, club racer, or autocrosser, this system will carry you a long way.


Teaser


Also free, made using 5hz data.

BT-Q818XT can be found at amazon for a good price. Phones can be found on ebay, newegg, etc.

Various photos, screen shots. Last photo courtesey of RaceChrono. Click for bigger picture.

 








RaceChrono Logo provided by RaceChrono
Samsung Blackjack photo is a Samsung publicity photo.


Monday, November 1, 2010

Gainesville Test Track Day Report - 10/30/10

My co-driver setting tire pressure before the event
Enjoyable weekend again. I had a co-driver today also, who fortunately didn't flip my car or destroy my transmission. His name is Brian, and I think you'll be seeing him on this forum more often.

Unfortunately, I do not think I will ever achieve Project OLDAZENIS.

I was only able to turn in a time of 1:02.49. No improvement since the last event. This is with me figuring out more parts of the course and benefiting from the added horsepower from the RX-7 AFM and a further 30lbs of weight savings. I had at least wanted to get in the 1:01.XX range, but the course was run in reverse for the second half of the day and I did not get to work on my driving.


But I have some good news. I had put together an endlink for really cheap. It is by far much cheaper than the other endlinks available on the market. It held up fine.

The RX-7 AFM gave me measurable gains. The previous Vmax (maximum velocity anywhere on the track) was at approximitely 72.6 ish, with my benchmark driver Grant driving the car. My previous Vmax is very close, but I believe it was a bit lower. Today, with temps in the low to mid 80's, my new Vmax was 73.6. This is probably partially for the 30lbs ish weight loss from removal of the passenger seat. To offset this influence, I was running almost a full tank of gas when I recorded this. Gasoline is approx 6lbs per gallon.

I also learned to squeeze out some more speed out of the course:
The insides of various have more grip. It might be cambered less. It might also be more abrasive, but I have yet to verify that. I am getting better at feeling out how much grip I have left on the rear wheels, helping me accelerate out of corners.

And finally, my driving continues to improve. I have to be smoother and keep better car control. This is just going to take practice. Every time my rear end slips out a bit I am losing speed. So if I do a very good lap, and then have the back end kick out or the car veer off line a bit, that lap will be wasted. I will keep shooting for consistency. I did have much more fast laps than last time, so my consistency must be improving.

Tire aging and grip
My driving keeps improving on this circuit. Because of this I find it difficult to gauge how fast my tires are wearing, and heat cycling because my driving is changing. My lines are changing, which makes the peak cornering areas different in each corner. Also, compared to my previous sessions, I am spending more time closer to the limit. However, the peak G forces are a bit higher on my old runs. Therefore, I conclude that my driving is getting better, and my tires are losing their grip. That is the only explanation I can come up with to explain the fact that my laptime has not improved, since I am driving better and my car is faster.


Results
Best time: 1:02.5
Temperatures about low 80's.

Changes in the car since last event
Custom RX-7 AFM mod
Passenger seat removed
Rear sway bar completely removed

Effect of removed passenger seat
With the passenger seat removed, the car cornered to the right much faster. Peak Gs are noticeably different, and I can feel it as I am driving also.

Driving improvements
I am being more consistent with my lines and hitting all of the apexes. However, I still come out too wide while entering turn 9. I should be closer to the inside. If I can be smooth consistently going into turn 4, I can hug the inside of the turn instead of being pushed out. No doubt faster. Please check out the videos at the bottom of this post.

My co-driver being chased down by a vintage formula Ford
Link to videos:
Jerry Lee 2nd session
Brian Inacay 2nd session

Link to pictures from this event