I receive a lot of emails and calls from people enquiring about different types and or years of cars and how each might be run with the original or a substitute engine. The following is an attempt to summarize a lot of those questions in a FAQ format.
This is a work in process so please be patient.
The very first question I ask people is "What are you interested in doing?". My responses are heavily dependant upon the goal.
For example, I am a mechanical and electrical nerd. I am curious about all aspects of the technology and would like to learn as much as I can about the engines, gearboxes, fuel controls, data acquisition, chassis set up, aero performance, baking, etc. This means I want to run period correct engines installed in the correct chassis and running all the nice aero bits close to how they ran in the day. I am also interested in going fast which is another reason to stick with the cars exactly as they ran in the day. These were half million dollar cars designed to be as competitive as they could be within the given set of rules. The solution for me is quite different than for someone who likes the look of these open wheel cars and wants to play with on track days.
You will need an answer to this question before the following sections will have much relevance.
What is the difference between the Cosworth XB and XD? The XD is a refinement of the XB including a lot of changes to allow the engines to rev to higher levels (roughly 13,200 for the XB and 14,700 for the XD). The cam drive was moved from the front of the engine to the rear, the intake system, throttling and heads were all redone and the electronics were improved. These are but a few of the changes made. The XD produces more horsepower.
One of the biggest technical differences between the engines is that the XB limits you to early 90s chassis (there were some back marker 96 teams running XBs) while the XD lets you run late 90s chassis.
More important than the technical differences to anyone interested in actually running the engine is that the XD is easier to support. Nicholson McLaren bought the XB bundle from Cosworth and developed a normally aspirated version of the motor. It is very difficult to get the proper original parts to refresh an XB. The XD project was just recently purchased from Cosworth by Ed Pink Racing Engines (2009) and they are committed to supporting the original turbocharged methanol burning engine run in ChampCar. They will sell you parts should you refresh engines yourself (like I do) or they will sell you complete engines and refresh them for you.
Your goals will dictate if the following applies to you but the response here has to do with how each year did the job of being a competitive chassis and how safety plays into the game. 2K and beyond chassis are not included as, at time of writing, the XFe engines were not available.
Throughout the early 90s the cars got faster. The move to carbon fiber accelerated that process and the sanctioning body stepped in hard at the end of 96 to calm lap times down. Tunnels were reduced and rear wings were regulated with the knowledge that smaller rear wings and tunnels mandated smaller front wings if you were to maintain balance. All the manufactures continued to strive for more mechanical grip and that process was largely unregulated. Given those two driving factors, it is reasonably safe to assume that the 96 chassis will be the best aero performer and the 99 should have the best mechanical grip. Given that these cars make their lap time from aero, the 96 would tend to be the "best" chassis.
From a safety perspective, the latter the chassis the better. Early 90s cars were aluminum monocoque with carbon uppers. My advice, do not go fast in an aluminum monocoque car. I sold my Chevy Spice for that reason. Carbon tubs do a much better job of surviving crashes. You can find a bunch of pre-94 Indy Car stuff out there in the $30,000 range. Just remember that they have a bunch of horsepower (800ish) and nowhere near the chassis should you get into trouble.
With respect to the carbon tub cars, the later the year, the better the safety. Tub volume was dramatically increased in 97. I am 6' 4" and 205 lbs and actually have a bead seat in my 99 as opposed to be neatly wedged into my 96 and actually needing to have the tub modified in my 95 to fit my shoulders. In addition, the 95/96 vintage Lolas had their front suspension lower A-arm pick up points attached to aluminum bulkheads inside the tub. The 97 through 99 chassis have external pick up points for all front a-arms.
From a practical standpoint, the turbo Buick engine cars never seem to run very long producing competitive horsepower and the early 90s DFs are bloody expensive to run, even more so than the XBs or XDs.
This goes back to the why do you want to do this question. A proper Indy or ChampCar has no place in an open track day. I have taken my car to historic weekends on their test day to break in engines and such but have never been able to run the cars hard unless it was (1) a proper HSR or SVRA weekend or (2) a private test day were I was piggy backing onto a pro team's test. These things are just too fast to get on the track with street cars or drivers without a reasonable amount of experience. The closing speeds are way too much for other guy to comprehend without a reasonable amount of experience. Lastly, these cars need heat in the tires to work which means you have to be leaning on them. You can not do that if you are trying to work your way around Vipers and Corvettes. You can not get the car to work and you end up looking like a fool.
If your looking for a fun track day car, you will be equally competitive and way safer with an Indy Lights car painted up in 94 Benneton livery. The horsepower to weight ratio is good enough to smoke all but the 1000 hp street cars and they can not even come close in the corners or under braking. In addition, the cost to buy and run the car will be far less.
The cheeky answer is.......
I prepare and run my cars myself. I have a group of people that ran the cars in the day (mostly x-Ganassi guys) that I can get to help me during the racing weekend but otherwise do most of the work myself.
In answer to the question, a long time crew chief once told me there is no magic. They are all just four corners, a gearbox and two lumps; one lump behind the fuel cell and the other behind the wheel. The two lumps are the expensive part.
The gearbox and corners are designed to be customer car parts capable of doing 500 mile races at an average of 220 MPH. I simply do not have any issues with corners have never had to replace bearings once I build the car the first time and have never found a crack. As for the gearbox, I will replace one or two gears a season for pitting at the root of a tooth and dog rings at the end of the year during the winter refresh. I'll replace pads every two to three races and rotors every four to five. I've got a large (very) stock of Brembo brake components so this is not expensive.
You can read more about the engines elsewhere in this site but I managed to get my XB engine program to twenty hours between refreshes and currently have my XD program to 12. I simply have not had enough in the way of refresh cycles to get the XD to twenty but feel it is possible. Springs are done every refresh and valves every other. Reducing the XD rpms from 14,700 to 13,000 dramatically increases service life of the block and rotating components. I use air gage tooling to check bearing wear and have NEVER had any measurable wear running Golden Spectral Oils. I run the engines 5 to 7% rich and was up to 50 hours on one of my XBs and ring seal was still good. Doing my own engines turns out to roughly $500 per hour for the XB and $800 for the XD. The XD number should come down as time between refresh goes up.
For reference, you can figure on $500 in entries, 1 1/2 on the engine, $1800 in tires (one set) and fuel for a weekend plus the cost of crew and transport.
The above costs are less than running an competitive Atlantic. It is the last lump behind the wheel that can really make things expensive :)
There are two sides to this question. The first and most important is that you can not see Methanol burning and thus it is more dangerous than gas. I use top end lube in the fuel to protect metal bits in the fuel cell so there is some smoke when mine burns but track marshals and the like need to be aware of what you are doing and know to use water. The nice thing about Methanol is that it takes very little water to kill its flammability (12% I think). This is why you see someone squirting a quick spray of water on an IndyCar after refueling.
The other aspect is the need to "pickle" the engine after you run it. I do this at the end of the weekend and it consists of hooking a small tank and pump (in a fishing tackle box) up to the fuel in and out on the engine, selecting the pickle map on the steering wheel map selector and running the engine for three to four minutes. This process puts gas into the engine's fuel system to keep the methanol from corroding the injectors and fuel rail. The pickle map setting reduces fuel delivery (less gas needed than methanol to run the engine) and fires the PCI ring injectors at low rpm which normally do not fire until later up the rpm range. The process takes fifteen minutes and I consider it a small price to pay for 800+ HP in a non-intercooled 1600lb car.