Nistune Forums

Yet another thread on the DLP A/D converter:

Setup:
Z31 w/ 88NA ECU, type 1, romulator
DLP Designs A/D converter (8channel)
AEM UEGO Wideband white wire plugged into input #1
*Verified WB voltage signal at A/D converter
*A/D converter grounded directly to ECU case
*WB grounded to Chassis

Sequence:
1) Plug A/D converter into laptop USB port, laptop recognizes hardware
2.) Open Nistune, set configuration
a. Under wideband type, select DLP A/D converter. Choose from
COM 2 or COM 3
b. Under wideband lookup table location, select
C:\Program Files\NIStune\Wideband\DLPADConverter, then
click okay
3.) Click WB 02 LH (or RH) button. Nistune attempts to connect.

Error message:

DLP converter can not be found on configured COM port.


Device manager shows the DLP converter plugged in (under USB), and the laptop recognizes it on connection. I've tried selecting COM 2 or COM 3 under wideband type (only two options) but neither works. I have tried first connecting consult. I have tried connecting, then starting nistune, or starting nistune then connecting.

Thanks,

-Mike

Okay.

I have solved my own problem, it was a driver problem. This may already be stated somewhere, but I didnt see it.

Here is how I installed:

1) Attach up to 8 0-5v input signals to the DLP board, Attach the ground to the vehicle chassis, and ensure each of the sensors your logging is also properly grounded to the chassis. Go to the bottom of this page and download the CDM drivers: http://www.dlpdesign.com/#Drivers

Plug in the DLP A/D converter into your computer. Windows will "find new hardware" and the installation/hardware wizard will appear. Choose to manually install (advanced, browse, etc etc) the drivers and choose the dlpdesign driver folder you just downloaded.


3) Open windows Device manager. Look for the USB serial converter, this is what windows calls the DLP designs board. The converter may already have a COM port assigned it. If not, You need to assign the DLP designs USB serial converter to a COM port.

Click on the USB serial converter (the DLP designs board) to open its properties. Within one of the properties, under one of the tabs there is a checkbox which says "load VCP". check this box. It allows the board to function as a virtual com port. Disconnect, reconnect, and possibly reinstall the above drivers. OR when you reconnect, it will show up under the device managers COM ports.

4) Open Nistune. Go to the configurations. Set the wideband type to DLP A/D Converter, select the COM port (as shown in the device manager), and choose the DLP folder under wideband look-up tables location. Press okay, click on the WB button, and it should connect. If the readings seem wrong, modify your lookup tables, and double check the grounds.

Good you found that. Normally VCP should be enabled by default, not sure why it was off ... but that would stop windows allocating a COM port for the converter if not ticked

We disable VCP on the FTDI245 chips but PLMS USB consult cables have VCP enabled to make a serial port for it

Just an update gentleman,

The W/B reading's on Nistune were not matching my UEGO gauge read-out. They appeared to be off by 1-2 AFR's.

To investigate, I monitored voltage output (white wire) of the AEM UEGO and compared it to the UEGO gauge read-out. Using a digital multimeter, I connected the white wire, and grounded to the same chassis location as the AEM is grounded too.

The voltage output of the white wire did not match the UEGO gauge read-out (according the AEM's supplied voltage-->AFR tables). I don't understand why. This appears to be a problem with the AEM UEGO, not a problem with Nistune or my connection.

What about using the serial output AFRs from the UEGO instead? Its meant to output these in 'text' format which we read in and display

So what you are saying is that either the voltage is not as expected, or the lookup tables are wrong

Further testing today.

I grabbed an extra automotive battery. I connected it to the AEM UEGO as its power source. I then unplugged the wideband o2 sensor, this defaults the gauge to 14.7 AFR, which should be around 2.3 volts (IIRC).

Using a Digital Multi-meter, voltage between the ground of the battery and the output (white wire) was 3.00 volts. This indicates a problem with the AEM UEGO. It is off by .7 volts, which is quite considerable (2 AFR or so).

Now, I could simply change my lookup table and offset each value by .7 volts. however, if the offset is incorrect, who is to say the gain is not incorrect also. Basically I need a custom lookup table for my janked gauge. Whos to say the display of the gauge is correct even. Perhaps these UEGO's are junk.

Matt. Your suggesting I use the blue wire. How would I do that? Is that a programming change you can make for me, or tell me how to make?

Lets say the white wires circuit is bad, but the blue wire circuit isnt. If thats the case, I could use the blue wire for my logging. But how can I check the blue wire to see if its good? It outputs data, not a voltage. I have no idea how to check what data its outputting, besides "use hyperterminal" which I have zero experience with. I will look around on how to hook this thing up to hyperterminal.

check on the other threads here. try getting the digital readout working. its your best bet at this stage

get a 9DB socket to plug into the serial port of your laptop/RS232-serial converter and solder the black/blue wires to this plug

check it works in hyperterminal firstly

nistune can read this output if you select UEGO from the wideband list

Matt,

That is the method I have read about, and I will be trying that method next.

IF the blue wire outputs incorrect data also (and I refrain my self from punting the AEM UEGO across the street) then I am going to make my own calibration curve:

1.) Record voltage with sensor unplugged (14.7 AFR)
2.) Remove sensor from exhaust, plug into guage, allow to settle (10 minutes) in free air. Record voltage as upper limit.
3.) Place sensor into gasoline soaked rag, inside tupperware container. Allow to settle, record voltage as lower limit.

Using these three values (and the knowledge that the AEM is a linear output), I should be able to fit a regression equation to the data.

What position is the calibration screw in on the back of the uego. It should be in either P1 or P2
Sounds like it could be in P3

Screw was in PO position. Which is same as P1 in terms of voltage output.

Update:

Tested AFR output using a Fluke 87 V meter, and a lab grade power supply to power the guage. Before I had been using a craftsmen non-auto ranging digital meter, and power through the car battery.

The fluke meter shows the correct voltage is being output from the gauge. I did have to correct a ground offset in my lookup tables, but everything is good now.

I just got my A/D converter in today and noticed the same thing with my AEM gauge was reading about 2+ AFR off from what Nistune was recording.

If I understand what you did correctly to fix this was to adjust the data chart used by Nistune to correct for a ground issue?

I have had also problems with the values from my AEM UEGO with the 0-5 Volt output over the DLP Board.
The values are leaner than the gauge itself. I tested the 0-5 Volt output as the the wideband sensor been unpluged.
The AEM manual say that with unpluged O2 sensor there must be 2,35 Volt at the 0-5 Volt output. I measured 2,45 Volt at my AEM UEGO.

I solved it by using the serial output from the AEM UEGO instead: The readings are rock stable, matches exactly the readings at the gauge itself and I got no more floundering values as with the 0-5 Volt output.

I would like to go with the serial port, but I have no laptop that has a serial input. And the past 3 serial to USB converters I've tried don't seem to work with my setup.

And yea, my numbers are leaning in Nistune then they are on the gauge. Would anyone know a good brand Serial->USB device that is know to work with Nistune, Vista, and an AEM setup?

Aten (Prolific USB chipset) or anything with FTDI chipset USB serial converters will work

Other cheaper chinese versions will not work well. I've had data corruption issues etc.

So has anyone figured out this problem, and have their computer matching the exact reading of their AEM wideband?