Help a n00b set up twin Webers: do I need wide-band O2 sensor(s) ???

Hi, guys.

I come to you for advice on basic Weber DCO tuning. I've been building a Mercedes 190E 2.3-16 for vintage racing over the last few years. The car will run in VARAC's Group 70+ class, which includes Group A touring cars from this era. Here's a link to the build thread:

https://grassrootsmotorsports.com/forum/build-projects-and-project-cars/87-mercedes-190e-23-16-vintage-racer-build-pic-hea/62702/page1/

We've just rebuilt the engine. This was the perfect opportunity to scrap the horrible (IMNSHO) Bosch CIS/KJet electromechanical fuel injection and single flapper-throttle setup in favor of twin Weber DCOS/P sidedrafts on a DTM manifold. Here's what we've got so far:

The engine is going into the car this week. The high-pressure (90 psi!) fuel system has been replaced by a low-pressure/high-flow pump and regulator from Weber/Redline. The venturis and jets have been sized by a Formula Atlantic engineer who has been helping with the build.

Hopefully it will run!

My question is: do I need one (or more) wideband O2 sensors to get the air:fuel ratio right? If so, I suppose I'd need some sort of data-logging to keep track of it. Would an EGT sensor/gauge be helpful?

My mechanic says he can get the car running with just a synchrometer and "by ear," then fine-tune by monitoring spark plugs. Is that good enough? He doesn't think O2 sensor(s) will really help much; he thinks it's just an unnecessary expense/hassle.

What do you guys think?

You can eyeball it, but you'll get better results faster with a wideband.

Your mechanic is wrong. There's good reason that NOBODY reads plugs anymore for air/fuel.

That's what I figured. Wideband O2 sensor makes sense to me.

That brings up some follow-up questions.

(1) How many O2 sensors will I need? There's a bung in the original downpipe. Is that good enough? Do I need a pair (one for each bank on the header)? One per throat?

(2) Is there a recommended brand? I'm thinking I'll probably need one that provides some sort of data acquisition.

In reply to Zomby Woof:

What might that good reason be? I read plugs all the time. It's the conventional wisdom about what those readings tell you that are wrong (hint: color is irrelevant).

I think an O2 sensor will just add confusion to setting up Webers. I'd use a dyno to really nail down the jetting.

You'd probably use one on the dyno anyway...

I think a single one should be good enough if you're setting the carbs up exactly the same and you're willing to leave the last 0.01% on the table that you might squeeze out of the engine if you could monitor the AFR on each cylinder individually.

Yes, people have tuned engines based on plug chops for a century, but people also used to mow their lawn with a Scythe. Why not make use of the more precise tool that allows you to achieve better results with hopefully fewer dyno pulls?

Factory bunghole (heh heh, bunghole) should be good enough, as BoxheadTim said just set the carbs up with equal jets and other tuning parts/settings. Unless one cylinder is known to run really hot or something (which I wouldn't expect from 80s MB) I doubt you're leaving much if anything on the table.

VClassics wrote: In reply to Zomby Woof: What might that good reason be?

Because:

a) few people really know how to do it

b) modern fuels make it even more difficult

c) there's no possible way you can match the speed and accuracy of a wideband

I don't know a single professional that reads plugs for A/F. Timing, maybe.

I didn't say plug chops are an effective way of jetting Webers (it would make you crazy), just that reading plugs is a legit diagnostic technique. When a misbehaving car comes to my shop, it's one of the first things I'm going to do to get a clue what's going on.

A morning on a dyno with a gas analyzer in the tail pipe is about the same price as a WB setup on the car (I've used Innovate), and will tell you more and more quickly. Also, if the car is going to run on leaded racing fuel, O2 sensors will be short-lived. If you need to retune for different altitudes or whatever, better to use EGTs -- find out what temps represent what AFRs on the dyno, then rejet to duplicate those.

So what do you guys recommend for a wideband O2 sensor? The Innovate MTX-L and AEM UEGO seem to be most popular. Both are priced at around $160.

I have the MTX-L, and it's fine. It seems Innovate have fixed their quality and reliability problems.

VClassics wrote: I didn't say plug chops are an effective way of jetting Webers (it would make you crazy), just that reading plugs is a legit diagnostic technique. When a misbehaving car comes to my shop, it's one of the first things I'm going to do to get a clue what's going on. A morning on a dyno with a gas analyzer in the tail pipe is about the same price as a WB setup on the car (I've used Innovate), and will tell you more and more quickly. Also, if the car is going to run on leaded racing fuel, O2 sensors will be short-lived. If you need to retune for different altitudes or whatever, better to use EGTs -- find out what temps represent what AFRs on the dyno, then rejet to duplicate those.

A gas analyzer may give you more information, but it's no more useful. As for speed- the WB's are way faster than the gas analyzers I've ever seen- the transport delay itself causes some odd data.

You mention altitude will require changes, if you get a WB, it stays with the car. So it can be done over. What dyno charges for 4 hours less than $200? Still- at the end of that morning, the analyzer stays. Vs driving all the time with the WB.

I'll take the sensor all day long, and next week, too.

Better if you can record engine speed and manifold vac as well- so one has an idea which part of the carbs need tuned....

I run Webers and I was taught to read the plugs. However, I also have a wideband which I use to confirm my settings.

I do indeed do the majority of Weber work by just paying attention to what the engine is saying.

It looks like your header is a 4-2 design. 1-4 and 2-3. So, there is no way to isolate each carb with its own wideband. Does your exhaust have a 2 to 1 collector further down?

If so, that's where to put the wideband.

When you choose a wideband, it is good to choose one that allows you to add TPS and RPM logging too. That way you can get a much better picture of what's going on.

Edit: TPS or MAP, as Alfadriver suggested. For Webers, I would opt for TPS though. The idea is the same, some kind of load reading.

I'm with ZW here.

TunaDad had a drag car, and sought, in the early days, to improve his tune on the dragstrip. He learned that driving back to the pits ruined whatever story the plugs had to tell. The only way to get an accurate look was the shut it off at the condition that he was looking to check (WoT for drag) and check the plugs on the top end of the return road.

So, for road racing under lots of throttle and load conditions, you are probably going to have a hard time tuning all of those points with just plugs.

Do you have one of the books for webers? The have a graph to help pick the intake size and from there it is simple math to pick the main jet.

The idle jets which are really mid range jets are a little more of a mystery.

One 02 sensor should work, since all the cylinders should be doing about the same thing.

The syc adjustment isn't really that complex. It's just a adjustment between the two carb's so they both flow the same, typically checked at or near idle.

A plug check won't tell you much, which jets do you change, main or idle?

With the 02 sensor, hold a steady speed at part throttle, that will show the idle jets. Then on the track, check steady speed at wide open or near wide open, that will show the main jet.

There is also a air jet that will move the main jet curve, and there might always be a change between the idle circuit and the main circuit.

Even a cheaper narrow band 02 sensor could be used, you might want a gauge on the dash, sometimes you might want to change the main jet by one size depending on conditions.

So why ditch EFI for carbs and still want to tune with a Wideband and datalog with a computer? My thought would be to sell the DCOEs and buy some Jenvey DCOE throttle bodies and run Megasquirt or some fancy/expensive ECU setup.

In reply to RossD:

Because you can? I may not want to drive a car with carbs, but many do. If you want to take data so that your carb tuning is better, nothing wrong with that.

For some, it's a whole lot easier to tune carbs than EFI. This does it better.

Well I think he is running some sort of Vintage race group. So you can use modern tech for tuning, but the car should be vintage set-up. Carb's work fine, maybe not as precise as modern injection, but there is a long history of race cars with all types of carb's

RossD wrote: So why ditch EFI for carbs and still want to tune with a Wideband and datalog with a computer? My thought would be to sell the DCOEs and buy some Jenvey DCOE throttle bodies and run Megasquirt or some fancy/expensive ECU setup.

The 2.3L 16v engine in my car was designed by Cosworth to make 265 hp in street trim with ITB's and Kugelfischer injection. Sadly, Mercedes decided to detune the engine for the production car: lower compression pistons, less aggressive cam profile, and (worst of all) Bosch CIS and a single throttle.

We've built a race engine based on the original design. The stock CIS system can't provide enough fuel for sustained high-RPM use, so it had to go. But it's a vintage racecar ... and ITB's plus modern EFI isn't really in the spirit of vintage racing.

Plus, I just think Webers are cool. Maybe I'll change my mind after I run this car a while LOL.

alfadriver wrote: If you want to take data so that your carb tuning is better, nothing wrong with that.

TED_fiestaHP wrote: Well I think he is running some sort of Vintage race group. So you can use modern tech for tuning, but the car should be vintage set-up.

Exactly! No reason not to tune the car properly just because I've chosen to run carbs.

LanEvo wrote: We've built a race engine based on the original design. The CIS system can't provide enough fuel for high-RPM use. But it's a vintage racecar ... and ITB's plus modern EFI isn't really in the spirit of vintage racing.

So my Alfa racer seems to have the same issue. Nobody wants to race SPICA, as they think it's not capable of running the fuel that it needs for more power.

But it appears your reality is just like mine- the real answer is that nobody really tried, unless you know the right people.

My friend actually ran an turbo on his Alfa using a SPCIA, and there's a kit out there that puts a turbo on a SPICA Alfa. Which tells me that the pump is more than capable of flowing the fuel.

A quick google check tells me that the CIS system was run on a turbo Porsche.

So the real problem is the set up. Just like for me.

Not trying to talk you out of the Webers- the most common fuel system for Alfa racers of all time. But pointing out that if you were really interested, and was willing to put some engineering in it- well, I bet it's possible. And it may make more power.

Had I stayed in racing, I was planning a fully tuned SPICA pump.

alfadriver wrote: ...the real answer is that nobody really tried, unless you know the right people...if you were really interested, and was willing to put some engineering in it- well, I bet it's possible. And it may make more power.

You make a good point. It's certainly theoretically possible to make CIS work, but I don't have the skills to do it. Over the past 5 years, I haven't been able to find anyone else who knows how to do it, either. Plus, parts prices are so high that I can't really experiment (I bought the pair of brand Weber DCOS/P 50's brand new for less than the price of a new CIS pump).

Besides, none of the 190E race cars ever ran with CIS and a single throttle. The setup I'm running is closer to that of the earliest 190E Group A cars. I just didn't see the point of trying to make it work.

Webers are just fun. I would recommend if you're running side drafts to get SK or OER carbs. They are Japanese copies of the Webers but, they are fully adjustable from outside the carb. I think you can change all the jets, accelerator pumps and squirters, and float level without opening up the carb which would make life a lot simpler. They use Weber jets.