Does a throttle bodie setup still need an afm?

[krijn]

I was wondering;

If you have a blacktop engine with original throttle bodies, is there still an airflowmeter inside?

Are there engine setups with throttle bodies and without an AFM?

If you would mount throttle bodies in for instance a redtop, can you make it work without an AFM and a programmable ECU (for instance megasquirt)?

[NoHachi]

You can use both with and without..though blacktops originally run off a map sensor setup. With these kinds of things its usually correct that if you have to ask, you're probably best of not tackling a project like this any time soon.

Luckily there is already some info available on the forums to get you up to speed: try searching on keywords like MAP, AFM and TPS, that should already come up with some usefull stuff and answer your questions.

[BenR]

I would never use an AFM.

In most cases running AlphaN (tps vs rpm) gives the smoothest drive and most instant response.

ECU's dont really like reading airflow on ITB's.

[NoHachi]

Running alphaN on a streetcar sucks..lots of tinkering, no chance of fuel efficiency and sensitive to any change in VE. Good for racecars though.
A large AFM hardly introduces any extra resistance and is still by far the most accurate way of measuring air inflow, so not necessarily inferior imho.
Best solution would probably still be running a TPS/MAP blend algorithm. MAP has exactly the same response as a normal TPS system (already using TPS dot for accel enrichment and other transients), so you just need the TPS controll for low vacuum conditions (off idle).

[BenR]

On a street car with a very conservative VE characterisitcs then MAP is good, its great for accounting all atmospheric conditons.

Even with alphaN blending at lower rpm's to stabalise for pulsing map values, i've yet to feel nice with a full Speed density setup. The nature of the MAP readings on ITB's once you are past but a few percent throttle becomes very small change for very large throttle change and the map basically becomes wasted unless you have a large resolution to play with.

On road going cars with ITB's we run mostly AlphaN and get great economy, much better throttle response, use the map sensor to correct for baro conditions.

Speed density takes alot of time setting up and mapping, and not many people are willing to pay for an ECU that will provide a smooth map with lots of breakpoints, aswell as the time spent mapping it to where i would be 100% happy.

Its just preference and most of our cars run so much overlap that you cant idle using map.

[NoHachi]

I understand your arguments Ben, but aren't you affraid of the effects of unforseen VE changes? Clogged cats/airfilters etc etc? Thats the reason I would like to run of off a map system for a daily driver..For a racecar - tps no contest.

There should be a way of working round any (more or less) constant VE changes with a TPS system..You should be able to compare and adjust VE tables by monitoring constant idle fueling and RPM. That would allow you to adjust the main MAP to the new VE allowing safe TPS mode..Time to learn C and buy a MS system. It worked nicely in GT power.

[BenR]

I am not a software engineer and have no interests in becomming one, so i will not be learning any coding nor waste my time with a budget megasquirt system. When your dealing with paying customers you cannoy present them with a horrible looking installation built around parts. Plus you cannot charge them for the time you spend dicking around building it. Not to mention the mapping software isnt very good.

Anyway, the ideas you present are valid, for OE manufacturers. Being a builder of performance engines people only come to me to go faster, in which cases problems such as blocked cats of air filters are non issues and in extreem cases would only see rich running conditions and pose no damage.

In our world for economy and basic short term fuel trim there is nothing wrong with low engine speed closed loop control so i dont need to have anything monitoring idle fueling and rpm.

If i was running a race car with a plenum setup then i would run a alphaN/speed density blend, of if we have vtec i would run speed density through out as we can net a stable signal strength at low rpm and idle.

[NoHachi]

Think you misinterpreted my last sentance (not so strange). I meant it was time for me to learn C and try it out. The MS would suit my budget and allow tinkering..not recommending it above anything else. Should probably have written that bit down more clearly.

[BenR]

what is 'c'?

sorry for any misunderstanding if there was any.

[Mux213]

Ben, its a programming language