BATC Forum

My Portsdown tx is coming along quite nicely and is about 80% assembled into its case containing easycap dongle and rtl rx dongle - a bit tight for space but it fits. Will post pictures shortly.

I manged to get this 3.5" display working nicely as a 35A waveshare http://www.ebay.co.uk/itm/3-5Inch-TFT-L ... 2749.l2649 good price and comes with a RPI case and some heatsinks.

Now here are my questions

1. The ADF4351 is not producing any RF output and the lock led isn't lit. I checked the control lines and only see traffic on them when going from rx to tx which presumably turns the chip off anyway. I see no traffic when I select tx or change freq. It is the 25Mhz version and is wired with straight wires (not twisted as this may cause a problem).
I have it powered from a 7V supply and the CE pin appears to be high so I guess it should work - any ideas as to whats going on and why I see no data when I select tx.

Edit = the board has 3.3V so likely not a supply problem.

2. At power up I get the BATC flash screen, how do I get it to go straight to menu mode ? I've read the docs but have yet to stumble on the solution.

3. When I select tx I get the BATC screen and have to guess where the button is to get back into rx. Is this normal ?

I'm using the latest image and have done the online update - maybe this has broken something with the ADF4351.

Besides the above irritants it's all looking very good !

73 Tim

Hi Tim

Pleased to hear that you're making progress, albeit with some problems to resolve. Here's some help:

1. Please check that you have IQ (Portsdown) selected as the output mode. The VCO does not get started in any other modes. (Console Option 2, then Option 3, then IQ with the spacebar).

2. To go straight into the touchscreen menu at boot, you need System Setup, Autostart and select Display Boot with the spacebar.

3. You should be able to get back to receive by touching the screen anywhere - I thought that I had fixed that (but maybe I haven't!). What is displayed during transmit depends on the video source. If you have the PiCam and are using CAMH264 or CAMMPEG2, you should see the transmitted image. You can deselect this in Advanced Systenm Setting, Viewfinder, which should leave you with the buttons displayed during transmit.

Please confirm your software version number?

Dave

More egg on face, another case of RTFM - having followed your instructions all is working nicely now which means I can package it all ready and only needs the RF switching once the shop opens.

Happy Days !

To answer your question, the code is 201705301 but with the latest updates applied.

Edit I'll do some more detailed testing tomorrow in various modes and different video sources etc.

Hi Tim

Well done. Thanks for letting me know!

Dave

Everything was working great yesterday then Murphy came out to play...

Bit of a long story but here goes....

Touchscreen stopped working completely - no display and no data from stylus position.

Rebuilt the code from scratch to the same recipe which worked and the touch screen began to work but no display.

Also noticed that the config.txt file was getting modified (corrupted?) some how as the edits I made had using nano had vanished.
Specifically to get the display to work I had to enable i2C as it was hashed out but something deleted it !

Try as I might to get the display back again nothing worked including the original vendor supplied image which worked out of the box) so concluded I must have toasted the display

Ordered another one from Amazon as per the wiki and in readiness rebuilt the code again to the Portsdown baseline - made no changes at all and just for kicks ran it up with only the callsign, locator , boot to lcd and shut down button enabled.

After the reboot everything sprang into life no need to mess with the config file or overlays etc - pi camera, display and touch all working great !!

Quickly made a back up of this image just in case.

Sadly after a few re-boots back to square one, no touch or display.

Time to use the backup image and try again - no joy again.

Starting to think it might be a speed / timing issue with the display as the PSU is a good stable 5.1V (even tried 5.2V)

I have the display mounted remotely to the RPI via 180mm of ribbon cable - as I've soldered the ribbon cable to the back of the display I can't directly mount the display on the RPI but will try with the new display when it arrives tomorrow.

However I thinking there might be something else going on as the shut down button doesn't always work - currently using Sandisk extreme 16GB memory to going to try Sandisk from BATC shop along with RF output switch PCB so time will tell if this is the root cause of the erratic behaviour).

While I've been writing this marathon e mail I've rebuilt the code from scratch again making only minimal changes as before - no LCD but touch worked once for a few seconds.

Looked in the config.txt and noticed that I2C was disabled, so I enabled it and rebooted. Display started to work then stopped.
Took another look at config.txt and I2C was disabled again - so I enabled it and rebooted but still no LCD

The more I think about this the more it looks like a timing problem possibly due to my use of a SanDisk extreme memory card and / or mounting the display remotely.

Wonder if it might make sense to reduce the speed of the I2C and SPI interfaces ?

Phew - that's all for now....

after a bit more digging I discovered that pin 23 on the LCD which is LCD clock was intermittent causing all sorts of display artefacts when it was sort of working.

A gentle pull on it removed it from the LCD PCB - no trace of solder on it at all, maybe a dry joint or perhaps they just rely on friction to make the contact !

I wonder how many other pins are afflicted the same way ?

Pushing it back in some what forcibly seems to have fixed all my problems (at least for now) - you live and learn !

Hi Tim

I've heard other users report that screen hardware problems have corrupted their SD Cards. Unfortunately this is a Raspberry Pi issue and there does not seem to be a workround.

Interesting to hear of the loose pin on the display. I wonder if this is the same issue as others have been having.

Dave

Hi Dave,

there is definitely something going on with the config.txt getting changed, but performing a 'proper' shut down seems to help.

As for the loose pin on the display it might be inherent with the connector pins effectively being inserted into a 'blind via' as the backlight / display prevents the pin going right thru' the PCB.

So I wouldn't be surprised if this was a major failure mode for the display.

73 Tim