• We recently switched our forum platform. If you experience any issues please email support@crystalfontz.com

CFA-635-TML-KL new user questions

mkwahler

New member
Hello all:

I've purchased a CFA-635-TML-KL LCD module, and having trouble getting it to work.

My final goal is to connect this module to a PIC18F microcontroller (not using any accessories, e.g. FSCAB, ATX; all I need is the display screen, buttons, and backlight), but I first tried to communicate with the module from a PC via the "cftest" program. I'm using a USB-serial converter (TTL-232RG-VIP-WE https://www.ftdichip.com/Support/Documents/DataSheets/Cables/DS_TTL-232RG_CABLES.pdf). This cable has a pin for the circuit to supply the voltage level to be used for signals. I've used this cable succesfully for many other applications (mostly with PIC MCUs).

First try:

Connections:
'H1' pin 1 - FTDI cable 'Tx' pin
'H1' pin 2 - FTDI cable 'Rx' pin
'H1' pin 15 - Ground
'H1' pin 16 - +5 VDC
+5 VDC - FTDI cable 'Vcc' pin

Questions:

- Startup behavior

When power is applied, there's a brief flash of a block of multicolored lines on the right-hand part of the screen (about a fourth of it), then it goes dark. (Also no LEDs are lit).

Should there be some 'boot screen' at startup, or is a dark screen normal? Should the LCD module be sending anything to the host when it starts?

- Baud rate

What is the correct default baud rate 'out of the box'?

Datasheet states (Section 9. Host Communications, page 21):
"The port settings are 115200 baud, 8 data bits, no parity, 1 stop bit by default. The speed can be set to 19200 baud under software control"

Yet later in the datasheet (9.4. Command Codes, page 41), I find:

"The factory default baud rate is 19200."

So which is it? Obviously, I won't get far without knowing correct baud rate to use.
(I did try both 19200 and 115200 with same result).

- Jumpers:

From the datasheet (3.2. Jumper Locations and Functions (All Interfaces), 3.3. Valid Jumper Settings), I determine that the interface type is selected with JP7, JP9/RJP9, and JP10/RJP10. My module, whose part no. ends with 'L', should be 'logic level serial.' My module was shipped with:

JP7 : open
JP9 : open
RJP9 : populated with 0-ohm resistor
JP10 : open
RJP10 : populated with 0-ohm resistor

Datasheet Section 3.3 (Page 12) contains text:

Function: USB interface, module powered by USB (not useful for -KL or -KS)
- JP7: Open
- JP9/RJP9: Closed
- JP10/RJP10: Closed

Function: Serial interface, module powered by H1 (default setting for -KL & -KS)
- JP7: Open
- JP9/RJP9: Closed
- JP10/RJP10: Open

So unless I misunderstand, this module seems to be configured for USB. There's no silk or label on the module indicating which interface, and I double-checked the invoice part no, which shows the one I wanted: CFA-635-TML-KL (logic level serial). Should these jumpers have been set for me, or am I expected to adjust jumpers myself? If the latter, that seems to obviate having a distinct part number for each interface type. Did I possibly get shipped the wrong unit?

Datasheet Section 2.2 (Page 11) contains text:

"The CFA635-xxx-KL has three jumpers. To open a jumper, remove the solder ball AND de-populate
the corresponding resistor. To close a jumper, melt a ball of solder across the gap OR populate the
corresponding resistor location with an 0603 0-ohm resistor."

This tells me that for JP9 and JP10, either soldering across the gap or installing the 0-ohm resistor will work. That also seems to confirm my suspicion that this module is jumpered for a USB interface.

I also did try wiring Tx/Rx directly to a PICs UART, and wrote code to light an LED when any character is received. I got essentially the same result: no LCD response (I did make sure my code worked by sending characters from a PC, and LED lit as expected.)

I suspect that perhaps I'm not undertanding correctly, or making invalid assumptions. I think I need better undertanding before I fire up the soldering iron and possibly dig myself deeper into trouble. Can anyone help me?

Thanks,
-Mike
Looking for additional LCD resources? Check out our LCD blog for the latest developments in LCD technology.
 
Hi,
I don't have the 635-KL, but looking at your info and the data sheets for the hardware, I will offer a couple of answers.

The data for TTL-232RG-VIP-WE cable indicates that the RED (Vcc) wire is an input expecting power from the customer's equipment (the 635). I suspect you think the TTL-232RG-VIP-WE will provide power to the 635, but it appears that's not the case. It would explain why the 635 flashes briefly and goes dark. Have you tried feeding +5 volts to H1-16 (without the TTL cable attached)? The 635 should boot and show a screen.

One other case may be the "RS232" logic levels may be inverted from what the cable/display needs. The 635 KL is described as "logic level, inverted" I/O. This is somewhat confusing, as an RS232 signal is inverted from the UART logic levels. Maybe you can test this on the H1-1 and H1-2 pins (cable disconnected) during bootup or at a keypress to see the voltage states; you might need a scope to see the waveform.

I would set JP7,10 open; JP9 closed. The data sheet says this is the KL default.

The schematic in the TTL-232RG-VIP-WE datasheet shows plenty of jumper options on its PCB. If you're adventurous, you might set JP1 and JP6 to feed USB +5 to the RED wire and solve that issue.
 

mkwahler

New member
Hi,
I don't have the 635-KL, but looking at your info and the data sheets for the hardware, I will offer a couple of answers.

The data for TTL-232RG-VIP-WE cable indicates that the RED (Vcc) wire is an input expecting power from the customer's equipment (the 635). I suspect you think the TTL-232RG-VIP-WE will provide power to the 635, but it appears that's not the case. It would explain why the 635 flashes briefly and goes dark. Have you tried feeding +5 volts to H1-16 (without the TTL cable attached)? The 635 should boot and show a screen.

One other case may be the "RS232" logic levels may be inverted from what the cable/display needs. The 635 KL is described as "logic level, inverted" I/O. This is somewhat confusing, as an RS232 signal is inverted from the UART logic levels. Maybe you can test this on the H1-1 and H1-2 pins (cable disconnected) during bootup or at a keypress to see the voltage states; you might need a scope to see the waveform.

I would set JP7,10 open; JP9 closed. The data sheet says this is the KL default.

The schematic in the TTL-232RG-VIP-WE datasheet shows plenty of jumper options on its PCB. If you're adventurous, you might set JP1 and JP6 to feed USB +5 to the RED wire and solve that issue.


Hello, thanks for responding. You wrote:

"I suspect you think the TTL-232RG-VIP-WE will provide power to the 635, but it appears that's not the case."

No, that's not what I expect, and I know for sure that it's not the case (clearly explained in the FTDI datasheet). As I wrote, I've used this cable successfully for many other applications (many years).
Perhaps I was not clear enough; I connected 5V (from a bench-top DC supply) to the H1 pins 15 & 16 (+5V to 16, ground to 15). I also connected 5V to the FTDI cable's voltage reference pin (red). (See the connections indicated in my first post, under "First try.")

You wrote: "The 635 should boot and show a screen."

So should this happen with only the power leads connected (H1 pins 15 and 16)? That definitely does not happen. Perhaps my module is damaged or defective?

About your comment per the 'inverted' IO, this might be something I need to look into, but I don't think that would prevent the module from starting, would it? And wouldn't I still see some kind of activity on the serial lines, even if the characters are wrong?

I just now applied 5V to the module power pins (15, 16), with Rx and Tx left disconnected. I still get the "brief flash of a block of multicolored lines on the right-hand part of the screen, then dark". The oscilloscope shows both Rx and Tx at a constant 3.3V (this doesn't seem right to me).

I forgot to mention in my previous post that I had tried pressing keys on the keypad, but didn't observe any activity on the serial lines (same result again today).

On your remarks about jumpers:
My module has both RJP9 and RJP10 populated with 0-ohm resistors (JP7 open). Yes, I also read in the datasheet that J7, JP10 and RJP10 open, and JP9/RJP9 closed (IOW only JP9/RJP9 closed) is the default for the KL model. But this is not how my module was recieved, which led to my question nabout whether I was shipped the correct module. My invoice shows the KL model (CFA635-TML-KL). I'm hesitant to desolder items from this unit until I have more definite information.

Anyone else have any ideas?

-Mike
 
Perhaps I was not clear enough; I connected 5V (from a bench-top DC supply) to the H1 pins 15 & 16 (+5V to 16, ground to 15). I also connected 5V to the FTDI cable's voltage reference pin (red). (See the connections indicated in my first post, under "First try.")
Yes, it was not clear to me that you used an external power supply. This hookup should work.

You wrote: "The 635 should boot and show a screen."

So should this happen with only the power leads connected (H1 pins 15 and 16)? That definitely does not happen. Perhaps my module is damaged or defective?
JP9 (or RJP9 installed) feeds H1-16 to the 3.3V regulator, which is the only connection needed to power the 635. JP10 is immaterial since the USB connector is not being used. So if the 635 doesn't boot up, it would seem to be defective.

... but I don't think that would prevent the module from starting, would it? And wouldn't I still see some kind of activity on the serial lines, even if the characters are wrong?
Yes, the I/O being inverted shouldn't affect bootup, AFAIK.

I just now applied 5V to the module power pins (15, 16), with Rx and Tx left disconnected. I still get the "brief flash of a block of multicolored lines on the right-hand part of the screen, then dark". The oscilloscope shows both Rx and Tx at a constant 3.3V (this doesn't seem right to me).
I agree. The 3.3V on H1-1 & H1-2 suggest that the 3.3V generator on the 635 is working.

My module has both RJP9 and RJP10 populated with 0-ohm resistors (JP7 open). Yes, I also read in the datasheet that J7, JP10 and RJP10 open, and JP9/RJP9 closed (IOW only JP9/RJP9 closed) is the default for the KL model. But this is not how my module was recieved, which led to my question nabout whether I was shipped the correct module.
I think JP10 (or RJP10 installed) makes no difference unless the USB connector is used, but I could be wrong.

I think you should send an email to support@crystalfontz.com or use the contact form, and give them a link to this thread. They should be able to straighten this out.
 

mkwahler

New member
Yes, it was not clear to me that you used an external power supply. This hookup should work.


JP9 (or RJP9 installed) feeds H1-16 to the 3.3V regulator, which is the only connection needed to power the 635. JP10 is immaterial since the USB connector is not being used. So if the 635 doesn't boot up, it would seem to be defective.


Yes, the I/O being inverted shouldn't affect bootup, AFAIK.


I agree. The 3.3V on H1-1 & H1-2 suggest that the 3.3V generator on the 635 is working.


I think JP10 (or RJP10 installed) makes no difference unless the USB connector is used, but I could be wrong.

I think you should send an email to support@crystalfontz.com or use the contact form, and give them a link to this thread. They should be able to straighten this out.
Thanks everyone for your input.

I've been in contact with Crystalfontz tech support, and we have indeed confirmed that there was a problem with the hardware. While waiting for shipment, I note that one of my questions here did not get answered (the one about default baud rate).
The datasheet has two conflicting statements about the default. One place it states 115200, in another, 19200 (See my original post for section/page nos.). Anyone know which it is? I've also asked this question on my support ticket, but didn't get a reply yet. Maybe you folks are faster. :)

Thanks again for the help.

-Mike
 
USB LCD Displays - Graphic and Character LCDs with a Keypad

CF Mark

Administrator
Hi Mike,

While waiting for shipment, I note that one of my questions here did not get answered (the one about default baud rate).
The datasheet has two conflicting statements about the default. One place it states 115200, in another, 19200 (See my original post for section/page nos.). Anyone know which it is?
The default is 115200.

Can I ask which datasheet version you're looking at so I can track this down?
It'll be at the top right of every page.

Ie:
CFA635-xxx-KU LCD Module
Datasheet Release Date 2017-04-19
Hardware v1.5 / Firmware u2v1

Thanks.
 

mkwahler

New member
Hi Mike,



The default is 115200.

Can I ask which datasheet version you're looking at so I can track this down?
It'll be at the top right of every page.

Ie:
CFA635-xxx-KU LCD Module
Datasheet Release Date 2017-04-19
Hardware v1.5 / Firmware u2v1

Thanks.
Hello Mark:

Thanks for responding.

The Datasheet I cited is "CFA635-xxx-KLDatasheetRelease2017-11-13.pdf"

Page header is:
CFA635-xxx-KL LCD Module
www.crystalfontz.com Datasheet Release Date 2017-11-13
Hardware v1.5 / Firmware s2.4


I also have "CFA635-xxx-KLDatasheetRelease2017-03-30.pdf"

CFA635-xxx-KL LCD Module
www.crystalfontz.com Datasheet Release Date 2017-03-30
Hardware v1.5 / Firmware s2v1


Both of these datasheets contain the same discrepancy. Is 2017-11-13 the latest?

Where these items are (pasted from my first post):
--------------------------------------------------------------
Datasheet states (Section 9. Host Communications, page 21):
"The port settings are 115200 baud, 8 data bits, no parity, 1 stop bit by default. The speed can be set to 19200 baud under software control"

Yet later in the datasheet (9.4. Command Codes, page 41), I find:
"The factory default baud rate is 19200."

So which is it?
--------------------------------------------------------------

Or just search for 'baud'

I don't remember if the module had any version info on it; I'm waiting for it to be returned by Crystalfontz.

While I have your attention: ( :) ). I have another question:

Does Crystalfontz sell a cable for this module? I did not find one on the web site. If not, and it's a standardized cable, can you tell me what it's called (IDC?), specifications, and perhaps refer a vendor?
Maybe this is better asked of your marketing people; if so, please forward my question.

Thanks,
-Mike
 

mkwahler

New member
Mike:

Sorry for the late reply.

The cable information is in the datasheet in section 7:

The following parts may be used to make a mating cable for H1:
• 16-position housing: Hirose DF11-16DS-2C / Digi-Key H2025-ND.
• Terminal: Hirose DF11-2428SC / Digi-Key H1504-ND.
• Pre-terminated interconnect wire: Hirose / Digi-Key H3BBT-10112-B4-ND (typical).

Hopefully that will get you going on the cable.
Hello,

Yes, I saw that info. But I'm wanting to buy, not build, cables for this module (CFA-635). Anticipated volume makes the time involved in building my own prohibitive.
I do understand that Crystalfontz does not offer a pre-made cable, which leads me to repeat my other question: Is this a 'standard' connector type, and if so, what is it called?
With that info, I can seek out a vendor.

To other users reading this, is anyone doing mass production (100's or 1000's)? If so, what is your cabling solution? Surely not custom building e.g. a thousand cables?

Thanks,
-Mike
 

CF Tech

Administrator
Mike:

Thank you for your reply.

We chose the Hirose 2mm series because there were many options for mating connectors (SMT, through hole, free hanging, etc) and it was polarized and friction latching.

Informally, we call it a Hirose 2mm, DF11 series 16 position, female. Between that and the part numbers above your cable vendor should be able to source it without trouble.

On this side, we are able to provide custom cables at the 1K level for very reasonable prices.

Please send a specification of the cable you need to support@crystalfontz.com and we can get a ticket open and a quote going.

Your sketch does not have to be super accurate or technical, you can take pictures of the mating parts for instance. As long as it conveys what is needed to make a cable we will formalize it in a counter drawing as part of the quote process.
 
Top