Quad-Channel, Analog Output Module - Review

Table of contents

RoadTest: Quad-Channel, Analog Output Module

Author: sadwallan

Creation date: 23 Aug 2017

Evaluation Type: Evaluation Boards

Did you receive all parts the manufacturer stated would be included in the package?: True

What other parts do you consider comparable to this product?: None

What were the biggest problems encountered?: 1. Missing cable 2. Unconnection or operation fail between the G.U.I and the device. Because of this two reasons, the desired final project was not able to be reached.

Detailed Review:

First of all, I would like to thank the RoadTest review system, the Element 14 team and Texas Instruments for giving me the opportunity to develop this roadtest. It has been my first roadtest developed, and I have to say that I hope that next ones will go better than this one, as I found two critical problems which limited me on going further on the developing of an application.

After saying that, I would explain all my procedure to test the device.

When I opened the box, I found the components of the following image:

First of all, I did a closer look into the given components of the RoadTest. What I found after a while is that the USB cable to interconnect the GUI with the SM-USB-DIG (programmer) was not the expected. They had sent a TYPE B to TYPE A cable instead of a TYPE A to TYPE A. Searching for the part number of the cable we obtained that the given was AK672/2-1, and the needed was AK670/2-2. I do not know why this error on the components sent, but it was clearly an error. Here I include a closer photo of the wrong cable.

However, I did not give up with this cable missing, as I found that it was an error but not impossible to solve on the way to test the device. In this way, I replaced the cable given by a new one of the correct type.

After changing the cable, I had installed the GUI provided for the DAC8775EVM, which was supposed to be compatible with the TIP215 (device to RoadTest). I have installed the GUI as a normal program, and it seem to be correctly installed. Here is the main menu of the GUI:

And here how the device was connected to the computer and the +24V PSU unit. The connection was made without the extender USB cable.

First of all, after connecting everything, I had tested the output voltage of the LM5166 of the board, and it was giving 12V from a 24VDC input. I have tested it by piercing one of the multimeter probes on the resistance R9 of the voltage regulator, and the other to GND, and it was operating normally. Moreover, the LED indicators on the USB interface were lighting, so I suppose that the behavior of all the device was correct, based on these two factors.

Then, I started modifying values on the DAC8775 interface in order to make it work. The first thing I noticed is that all the registers seemed to work well, as when a data value was written onto any register, making a read of this register returned the value inputted. In this way, I discarded a wrong communication between the device and the computer. Here is the sub-menu of the GUI that I used to check this.

However, here finished my luck. When I started the DAC Output A, pressing the button Output Enable without touching anything more, a 3.14V value appeared on the DAC Output A. The GUI configuration was the following one:

As it can be seen, with the configuration given, the DAC Output A must be 0V. In any case should be 3.14V, as this value is not expected by any data values.

Moreover, I have found more incorrect values on the different channels of the device. When going to the DAC Output B, and doing the same procedure that in DAC Output A, I have obtained a voltage output value of 0.64V. The other two channels had the same effect, with different voltages.

Finally, I changed different Buck-Boost modes and enables in order to see if anything was modified. Actually, I was not able to obtain any output voltage relative to the DATA inputted. I tried all the combinations I could think in all the channels, but surprisingly I was unable to make it work nor obtain any output voltage relative to the DATA values.

To check all the voltage values I have used a Kaise DT9201A DMM, which worked as expected when measuring the +24VDC PSU unit.

Concluding:

First of all, I have to say that I have found little information on how the GUI works. It has resulted on a slightly difficult behavior on try and error, but without receiving any good response. Probably I have missed anything, but I think that I have focused onto the device behavior as further as I could.

About the board, I have to say that seems robust and fiable, but I do not like the testing connector so much. I find it not applicable for the industrial field (I suppose that this connector is here because of the EVM). The GND field and the GND paths seem correct, isolating the DC-DC converter GND from the other GND signals. The rubber pieces in the rear of the board are really useful. Finally, the connectors J1, J2, J4, J5 and J6 are correct for the given application.

About the RoadTest, I have to say that it made me a little bit disappointed, as I expected a enjoying experience, but it have become kind of... sad development. I have not enjoyed so much the roadtesting, and in the end, I do not have any conclusion on the behavior of the board to use it in future applications. I hope that the rest of the RoadTesters will arrive to any conclusion, where I was unable to arrive.

Thanks for reading, any suggestion will be great to apply.

Carlos March.

UPDATE at 03/10:

After checking Peter Oakes review on this EVM (Quad-Channel, Analog Output Module - Review ) I had tried again if I was able to make the board behavior as expected. Unfortunately, I missed an essential thing, that was the establishment of the reference output voltage... I have to say that now, the board is working properly on all the tested modes. It is quite a shame to have missed a so essential point on the testing of the board, but I feel kind of good now that I have been able to solve the problem, and in this way, continue with the project idea. I have updated the score of this roadtest, more accordingly to the final review with the working EVM.