DBW Throttle Body Options

[samgm2]

After a great deal of testing, I've sadly come to the conclusion that using an 03 - 05 DBW computer to control an LS2/3/7 style OEM or Aftermarket Throttle Body is not straightforward.

The 03 - 05 OEM Throttle Bodies output from low to high voltage and are essentially identical output (or should be).

The LS2/LS3/LS7 OEM Throttle Bodies use an old-school potentiometer design from the dark ages. There are two sensor outputs. Sensor 1 will basically output from low to high voltage and sensor 2 will output from High to low voltage.

Aftermarket Chinese sensors use a far more modern approach utilizing a diametrically magnetized magnet axially oriented over a TDK HAR-3725 MEMS sensor. Although much simpler to implement, the design is far more robust and much safer than the OEM design. The sensor chip itself is fully redundant and in combination with the computer will always FAIL-SAFE. The part allows for flashing (up to 100 times). Although a bit obtuse in its register implementation, it also allows for an arbitrary waveform output allowing for both gain and offset segments in order to linearize the output. Amazingly, it has options for DAC and PWM outputs. All of this is through simple register configuration; nice part.

I used a similar part from Infineon (TLV493) and obtained amazing accuracy with the device (better than 0.1%) but that required a separate microcontroller (I used an XMC-4200 for that project). In those projects, I used the sensor to determine both rotation and magnetic field shape characterization.

The TDK part doesn't require ANY support components, just resistors, and capacitors making it much easier to design a circuit with. I should be able to generate a similar response to the old OEM sensor but using a much better approach.

Part of me thinks this might be a worthwhile project. There seems to be no option out there for running a Vortec-type truck computer with a newer style TB. This would address that problem and allow a simple swap of the PCB for a cheap AFTERMARKET TB and eliminate wire harness adapters and ridiculously expensive voltage translators.

Thoughts? Have I missed a simpler solution out there? This one requires a PCB design as well as some programming to custom configure the TDK HAR3725.

[SiriusC1024]

Yeah. It's called a cable and segment swap.

Specialist in Advanced Teledildonics.

[GHuggins]

Nick Williams offers TB's for these. Just have to make sure your ordering right. Might even have to call his shop to get the right one. Literally just finished tuning an 04 vette running a NW103 that I know Nick swapped the motor and tps to make it right for the guy.

[atkinsond]

http://www.torquerush.com/x-link.html

[samgm2]

My cable is correct, which PCM/vehicle would you segment swap with? My PCM is a 2004 Silverado. Which compatible PCM utilizes an LS2/3/7 throttle body?

[kevin87turbot]

My cable is correct, which PCM/vehicle would you segment swap with? My PCM is a 2004 Silverado.

Sirius1024 was referring to swapping to a DBC TB on those applications. We've done several of the P59 DBC conversions at my shop. The Express Van and V6 PCMs work great for that. I do like your idea of offering a true plug and play option though. It would be nice to have one that works properly without the X-link.

Greg, Does the NW version require the X-Link?

[samgm2]

Atkinsond, the X-Link is far too expensive to be a real option. Replacing the circuit or the IC in an aftermarket TB would be reliable as well as cost-effective.

[samgm2]

Kevin87Turbot, A DBC swap requires that I get rid of my PCM in lieu of a PCM with an installed IAC controller. I would lose the license on my current - now useless PCM. Finally, it would require me to Kludge a DBC cable onto my engine - an outdated old-school solution. DBW offers real advantages over DBC. Additionally, aftermarket and inexpensive TBs are actually superior to failure-prone resistive TBs that can end up with dead zones.

A magnetic solution utilizing a 3D digital hall effect sensor is superior not just in performance and linearity but will never suffer degradation. But again, if there is a solution that doesn't require me to design a replacement PCB, and software for it, I would love to consider it.

[SiriusC1024]

How can it be linear? The B-field isn't linear. You'd need a transfer function. For that you'd have to verify accuracy with a secondary sensor. Maybe you're confusing precision with accuracy?

Or are you wanting to make a "translator" circuit from say PWM out to the ECU?

Cable bro. Cable.

[GHuggins]

Sirius1024 was referring to swapping to a DBC TB on those applications. We've done several of the P59 DBC conversions at my shop. The Express Van and V6 PCMs work great for that. I do like your idea of offering a true plug and play option though. It would be nice to have one that works properly without the X-link.

Greg, Does the NW version require the X-Link?

No, but I do know Nick provided the guy with the correct internals to run it (tps and motor). Connector on the vette plugged straight into the TB. I remember looking at the wiring and it was rather aged - didn't see anything in between the factory and the tb - not to say there possibly wasn't something there? PITA to tune with the huge friggin cam he had, but I finally got it nearly driving like a stock car. His was making a little over 600 to the tires for whatever it's worth if that helps any.

[Alvin]

Atkinsond, the X-Link is far too expensive to be a real option. Replacing the circuit or the IC in an aftermarket TB would be reliable as well as cost-effective.

Re-inventing that will likely cost you more than just buying it if you value your time at all. Not meant to be a dig but making sure the TAC module is happy is going to be a pain.

I've not had any problems with the x links. I have one on our shop truck for more than 10 years.


I don't see anything about the setup... As someone who works 9-5 on this stuff with a chassis dyno.. A larger throttle body on a NA setup rarely offers much in terms of gains. Most people are fooled by throttle differences between a larger and stock throttle body.

I honestly can't say I've seen anything pickup even 10 rwhp. Most stock LS3/7 throttlebody to 103mm throttle bodies on even radical NA 500 RWHP + builds might see 2-4 rwhp if that. Just to give you reference.


PD blowers like magnacharger, LSA, whipple,etc. Those are the setups that really benefit from this.

[samgm2]

SiriusC1024, of course you have to apply some math. The part (HAR3725) performs the linearization. Additionally, it allows for the addition of 32 line segments to fine-tune that linearization. It's a really nice part. I can appreciate the desire to "simplify" the system but I don't like cables. Control is a beautiful thing.

[samgm2]

You make some good points, Alvin. And yes, I will spend a lot more than the cost of the x-link to get this done. I just want to try it. It will be a low-priority project to be sure. I just don't think it will take me very long. I will throw it into the next batch of boards that I have made. In the meantime, I will throw a plate adapter on to fit the old throttle body so I don't have to swap intakes.

[Alvin]

You can get PCBs made 10 for $2 now.

I don't save up any design projects anymore.. send it.