TMDSSOLARPEXPKIT: F28035 MCU control card

A closer look at the problem seems to imply that the controller care is powered, but not running the code.

Hi Roger,

Try flashing the code again. Here's the code location:
C:\ti\controlSUITE\development_kits\SolarExplorer_v1.1

Regards,
Gautam

I found the code. Do you think the code for both controllers needs to be reloaded? Is there a way to check that I have the correct code and it will run correctly?

Roger Bell said:

Is there a way to check that I have the correct code and it will run correctly?

Simply check the hardware version and respective folder in controlSuite as linked above in my previous post.

Regards,

Gautam

Roger,

Please follow the UG ,

if you want to just run the demo then you need to follow

C:\ti\controlSUITE\development_kits\SolarExplorer_v1.1\~GUI

otherwise you can follow the

C:\ti\controlSUITE\development_kits\SolarExplorer_v1.1\~Docs

for the full project

the PV emulator should have been programmed already and you do not need to download code into it..

the QSG above will program the control card on the DIM100 socket with the correct code.

Hi Manish,

The PC and the kit are not connecting. I have CCS 6.0.1 and SolarExplorer_v1.0.  I am a new to CCS with little experience.  I have made Leds blink on other kits, F2837xD Workshop is en example.  I used the same Workshop as an example on how to set up the Target Configuration for the SolarExplorer.  Could you give me some help on getting the PC and Kit to talk?

CCS was able to connect to the F28027 used to control the solar emulator.  I only connected to the controller without changing any of the code. I am at a loss as to how to resolve the connection to the controller card.

Common culprits

1. [Main]-J4 & [M7]-J4 are not populated (see section 3 C:\ti\controlSUITE\development_kits\SolarExplorer_v1.1\SolarExplorer_PVInverter_F2803x\~Docs)
2. One of the control card settings is incorrect
<same doc as reference above>
Do the following switch settings on the controlCARD:
a. Control Card SW1 is in the OFF position
b. Control Card SW2, Position 1 = ON, Position 2 = ON

Fig 13

3. Do not connect USB to the F28027 on the board, as both will enumerate as emulators and then CCS will be confused.

Follow the steps in the UG above and point exactly at what step things don't work as outlined in the UG.

(Sorry for pointing to the UG again and again, but for the issue to be identified i need to understand what steps were followed to conclude that things are not working)

Thanks for your time and patients with this issue. I am making every effort to use the documents and other online resources to solve this issue.

A little about my back ground. I am new to digital power, but have about 30 years of switching power design. Much of my experience is in Aerospace. I did taking a class by Biricha in Huston some time back. When the digital design of a system does not function as I expect, I find it hard to know how to solve the issue, i.e. the PC will not communicate with the controller.

I will verify that the switches are set as out lined. I have checked the switch setting and will check them again. If the switches are set incorrectly does the system require a power down to recognize the new setting?

The documents I am using are from v1.0. I am downloading CCSv7.0. It will take me all day as my web connection is slow. I did check the jumpers and switches. All are set as you outlined. When I use CCSv6 connected to [M7] JP1 and test the communication using Target Configurations set button, the test fails. When I connected the PC to the USB to the F28027 on the board, the test passed. The F28027 connection is not used during any other testing.

Outline of steps as of today
1. Powered the system not connected to a PC. Have an electronic load connected to the Solar Simulator, [Main] BS1-2. The output is low about 5.5V at no load. At no load this voltage should be close to 27V. The Solar Simulator does not respond to the light sensor. The output of the light sensor does change with light intensity.

2. It seems that the code in the controller card is not running or not correct. Thus, I started the process of reloading the code to the controller. At the time I found the PC will not communicate to the controller card. I would thank there are switches that are not correct. I have checked the setting and they seem to be set as required. Would it be helpful for me to send photos of the setup?

3. After about 3 sec of turning on the dc house keeping power switch the relay enable goes high. Does this imply that the code in the controller card is running? What does this imply about the PC communication?

Is the next step to get CCS to communicate to the system using [M7] JP1?

The PC is now communicating with the Solar Explorer Kit. I had to add a jumper to [Mail] J4. I did load CCSv7 to my PC. Should the code for the Solar Explorer Kit be part of the load? The project I found is not complete.

Roger,

I am glad that you are able to atleast load code into the control card.

there are two ways in which the design can be evaluated one is a QSG mode where the GUI will load code for you into the controller.

However as you have been having issues running the system as is, i would recommend following the steps in the UG (C:\ti\controlSUITE\development_kits\SolarExplorer_v1.1\SolarExplorer_PVInverter_F2803x\~Docs)

In which you will have step by step build which help in resolving issues quickly.

That would involve you loading code into the control card.

I would like some help. Will the PV Simulator function without connecting to CCS?

Thanks. I have a major issue now. The 3.3v is failing. The voltage is 4V+. I will work this voltage issue. Additionally, the 20V switch failed closed. These kind of failures imply there is a hardware issue, a problem with the kit or how I am using it.

Where can I get SolarExplorer_v1.1? I have SolarExplorer_v1.0. I did load CCSv7, but did not get SolarExplorer_v1.1.

Thanks, I will have to replace the 20V switch as it is shorted on. I found the Docs referenced above. It will be two weeks before much is done, as I can not work the issue full time this next week.

Hi Manish,

I did soom testing of the PVEMU section of the kit.  The PVEMU does produce a voltage and the voltage decreases with load.  This is as expected.  The issue is the voltage is very low, about 5.6V.  When I decreased the feed back voltage by about half the output voltage increased by about twice.  If this was a desecrate design, I would look at the reference voltage expecting it to low.  How do I look at the  reference voltage on the F28027 used to control the PVEMU?  Is it possible to look at the code without reloading the code to the F28027?

The box for "Notify me when someone replies to this post" is checked, but I am not get the emails.

Hi Manish,
I plan to check each section in the kit one at a time. I started by testing the PVEMU. The output voltage is low, about 5.5V. With an external load connected to the output voltage, the voltage decreases as the load is increased. The decreasing voltage implies the controller is functioning similar to what is expected with a PV panel. At no load one would expect the PVEMU to be at Vmax. The feed back divider is 100K, 11K, with 3V as the reference, the output should be about 27.3V. I decreased the 11k by about half and the output increased by about twice. Thus, the circuit is tracking the reference from the controller. This is true with no load or a load, max load used was 2A.

The controller uses a lookup table to set the reference. The load current sets the reference to given the correct voltage. To get greater than 20V the converter operates in the boost mode. I am not sure how the controller changes from buck to boost. It seems to me the place to start to find the issue is to look at the reference voltage and compare this value with the table.

The following is from the lookup table:
.sect "PV_TABLE_SECT"
; U Val (Q9); Index I Value U Value
_PV_Table1 .word 14336 ; 0 0.000000 28.000000
.word 14326 ; 1 0.005859 27.980923

I am not sure what the 14336 is? For that max voltage the reference would be 3.0V, for a 12 bit AD the value would be 4096. What am i missing?

To check the value used by the controller as the reference, I have to load the F28027 with the code using CCS. Is it possible to check the value in code that is running in flash?

The VP EMU is not operating as expected. The output voltage at no load should be about 27V not 5.6V. I would like to check the reference voltage the controller is using. Is this possible without reloading the code to the F28027?

I reloaded the f28027 used for the solar EMU. It functions the same as before the code was loaded. It seems that the table used to determine the output voltage is not being run by the code. How can I check the table is used as required?

I completed through step4 with the expected results. At stop 5 the panel voltage is 0 with 100 ohms or no load. Vin_sw is 19.69V. D2 is blinking and will stop blinking if the code stopped using the suspend button.

I am not getting email notification on my posting. How can this be changed?

I found an error the the code change. The voltage is 14.0V at no load and 13.9V at 100 ohm load.

At Luminance equal to 0.4, panel voltage is 11.08V at 100 ohms, 11.23V at no load.

I am glad at least you are getting some results..

considering this email thread is so long i would recommend you post the new question as a separate thread specific to the new issue you have, it has become very hard for me to track this on multiple pages..