SINUS RDS PROJECT

[sinus trouble]

Hello Necks

I have been taking some measurements today and came up with some interesting findings!

Firstly some not so interesting news, None of the unused pins have anything going on!
However pin 10 has some kind of data string present.
Pin 11 seems to have a steady square wave present around 1Khz

Onto the output itself, I captured it just to show what to expect from a correctly functioning unit.

RDS OUTPUT.PNG

[sinus trouble]

Now to the interesting part! The Data!

I captured both the output from a USB Serial device VS a REAL Serial socket.

USB CAPTURE.PNG

[sinus trouble]

SERIAL CAPTURE.PNG

[sinus trouble]

As a comparison, The data string itself is identical but inverted?

Also an obvious difference in voltage! Q3 and R12 on my schematic will protect the PIC from seeing anything higher than 5Volt but will also invert the data.

My theory is that the RDS could possibly be programmed Via USB by omitting Q3 and R12 direct to the PIC?

I will put this to the test very soon.

[radium98]

My theory is that the RDS could possibly be programmed Via USB by omitting Q3 and R12 direct to the PIC?

did you try that ? sinus

[sinus trouble]

did you try that ? sinus

I did indeed Radium! And i can now confirm that i have successfully programmed the 'SINUS RDS' via USB!

When i removed Q3 and R12 The RDS Encoder started behaving erratic! The data write LED (red) would stay on or flicker constantly.

So i re-fitted a 10K resistor in place of R12 to act as a pull up!

Now it works everytime, Not a single glitch!!

[radium98]

Nice 1 Sinus trouble very happy for that ,what abt a diy usb to serial that can work with a JDM programmer ?

[sinus trouble]

Thank you Radium!

I am not familiar with a JDM? A DIY USB/SERIAL is not quite as simple as it seems? I believe the cable contains an IC built in to handle the conversion?

Then you have the problem of driver software which most USB devices require to operate correctly.

RS-232 Ports are pretty much configured and ready to go.

Anyone who has a much better understanding of operating systems than me, please feel free to correct me if im wrong?

[Albert H]

The internal RS232 on most desktop computers gives -12V and +12V outputs (usually generated with a clock 'n' pump chip like the MAX232 from the 5V supply). Unfortunately, most laptops only give 5V and 0V for their attempt at RS232 - I used to build an adaptor inside a 9-way connector which took the weedy signal from the laptop and pumped it up to the correct voltages with a MAX232. This required a separate power supply to give enough "grunt" for the RS232 to be properly robust.

The JDM PIC programmer is a crude thing that uses four diodes, two zeners, a couple of generic NPN transistors an LED, three resistors and a ZIF socket. It needs the full fat RS232 to work properly.

The USB to RS232 adaptors usually use the Prolific PL2303 IC, and give the same weedy output as a laptop. The Linux kernel includes a driver for this IC, so it's trivial to get going in Linux. There is also a really nifty bit of software called GnuPIC that will drive most programmers. It's close to the full functionality of MPASM, and there are also interpreters for BASIC, FORTH, PERL and C, to allow you to write "high level" programmes and run them on a PIC.

There was a nice little article in Elektor some years ago that explained how to build an RS232 buffer box to beef up your laptop output to proper RS232 levels. If I can find it, I'll put it up here.

Another PIC programmer tool that I like (and can even be run under Windoze) is "PICPgm". The guy that wrote it also sells a fairly cheap programmer kit for a USB-driven programmer. His software supports lots of other programmers too - JDM, Tait, Olimex, El Cheapo, EPIC (the one I mostly use) and lots of others.

Windoze driver software for the Prolific chip was OK in XP, worked (mostly) in Win7 and was buggered in subsequent versions. You might be better off writing your own or searching the 'net for drivers that actually work. M$ seem to think that RS232 doesn't matter any more, so don't bother to support it!

[radium98]

thank you Albert i use picpgm last one very good ,my desk do not have the com port ,it is not that new model ,old abt10 years ,but sadly do not have the port ,i tried to take the output with the profilic no copy and yes i face the low output voltage ,and you are right once i have read that i need an extra voltage but also i forget where ,so every time i have to burn a hex or run to control an rds i have to put an old pc with their cable for just some little time ,anyway thank you.and thanks Sinus ,so why i told you abt that for the control of the rds.,i still have around a pci socket new that have pallel LPT andcom port.for the diy usb yes it need a chip that to be built it cost twice the price of a converter itself comonknown chip ftdi .but useles for the jdm .

[Albert H]

I've found the RS232 Booster circuit (and PCB layout). I usually build them on Veroboard (I'll find my layout and put it up here soon). In the meantime, if your computer, laptop, Raspberry Pi or USB adaptor only gives 5V for its "RS232", this booster sorts out the problem:

RS232-Booster.png

All the little electrolytic capacitors are 1µF @25V. I also bypass the 78L05 with 47nF ceramic caps right next to the pins to keep the chip stable in the presence of RF.

[radium98]

thank you Albert H a lot

[radium98]

Albert H sorry can you tell about the Elektor issue number please .

http://electronics-diy.com/electronic_s ... php?id=778

[sinus trouble]

Thank you both for the explanations, Its all beginning to make sense now!

I have found that the voltage as such is not an issue? The RDS is capable of receiving the data @ 5V or even as low as 3.3V.

The problem is the Inversion of the data? (Ref to GND 0V) The data write LED (Red) responds to the USB data but no change in RDS TEXT.

As to preserve the SINUS RDS as it is and avoid too many modifications i may build a very basic inverter PCB?
Hopefully i can get it to run from the USB 5V DC too.

Also just so theres no confusion, The USB cable i am using is a TTL-232 or F type (I think) i will upload a pic soon.

[sinus trouble]

20201210_233700.jpg

[sinus trouble]

Pin out
Red = +5V
Black = 0V
Green = Data

[radium98]

Sinus try with a simple ic 7404 hex 8 inverters between stages

[Albert H]

Sinus
The reason that I pointed out that the RS232 booster/adaptor is necessary is for many of the serial PIC programmers. Most of the simpler programmers use the voltages developed from the RS232 port in the host computer. If you're using a laptop or other hardware that just does a weedy impression of RS232, you'll be unable to programme most of your PICs.

The data input to your RDS board will accept TTL level fed straight to the PIC, and it's fortunate that the Prolific USB - "RS232" chip inverts the data! That's why you can talk to the RDS board with your USB adaptor lead.

My old RDS board (the one that Roger used to sell) used a USB IC that gave TTL levels out and talked straight to the PIC. This made for a more "generic" solution, particularly as more computers are equipped with USB than RS232 these days.

[sinus trouble]

I cannot disagree Albert!

These 'Adaptors' may work in certain applications? The claim by manufacturers that these items are 'True' RS-232 substitutes is simply misleading information!

I guess i am lucky to have the gear to put them to the test.

No doubt real RS-232 is my preference! USB is insignificant to me, just a possible option.

[Albert H]

I've been messing with the modulator end of the RDS coder. This is what I came up with in the middle of last night:

RDS Proto Latest.jpg

I'm running the LM358 from a dual supply (+/- 9V), but I've seen an easy way of adding some bias resistors so that it can run on the 12V supply or even the 5V supply.

The balanced diode-switching modulator is well-known in amateur radio - so I thought that I'd try it for the 57kHz modulator here. The results are a really robust RDS signal, with no nasty sidebands. I used 220µH inductors because I have a lot of them, and with 33n and 2n2 in parallel, it's really close to 57 kHz. It's worth selecting the capacitors (and checking the accuracy of the inductor) to make each tuned circuit resonate exactly on 57 kHz. I used the LM358 because they work rail-to-rail, have reasonably low noise and distortion, and because I have a whole lot of them! You could use almost any low noise dual op-amp (don't use a 4558, 748, or any other ancient part though).

Incidentally, there are a couple of bugs in the .hex for this project. I've been disassembling the code and I think that I know what's wrong. The bug causes momentary losses of the RDS data, leading to a blinking RDS indicator on some radios (my Sony portable ignores the problem, but my Technics tuner falls back to showing the frequency now and then). When I've cracked the hex issue, I'll put an improved version up on here.

The two-chip PLL allows synchronisation with stereo coders, but isn't entirely without issues. I still think replacing the clock crystal on the PIC is going to prove to be more robust.

I'm not entirely sure that the diode switching modulator is the ultimate solution, but it's going in the right direction. It's going to increase the cost and complexity of the coder, but will give better results.