Playing With SI5351B & Trying to Produce Voice FM Modulation

[@M0UAW_IO83]

Nice work, I did similar with the 'B' chip too, but approached it in a slightly different way On the '059 chassis you can replace the VCO with the 5351A and use the original modulation circuit (which aint great on the '059 but can be improved) and just modulate one of the loop mixer crystals for FM.

[@M0XFXUK]

Hi yes the B version is quite an interesting IC. since making this video I have made some significant improvements, I made a complete board with an ATMEGA328P controller and op-amps with pre-emphasis and signal limiting to the si5351B PLL. The FM is now sounding much better. I will make another video when I find the time and go through it. Interesting stuff regarding the 059 chassis.

[@thomass.knutsen2487]

Due to how the input of the chip is done, you need to bias the input of the VC control at something, not letting it float like in your example. This is the reason for your distortion and issues with large deviations.

[@thomass.knutsen2487]

a voltage divider consisting of 2 equal resistors between VCC and GND works niceley.

[@M0XFXUK]

Hi Thomas thank you for your comment, and yes I did figure that out in the end. I am now using 2 x 100K resistors as biasing to the VC pin between ground and VCC with an AC coupling cap, I have also added limiting and pre emphasis it's working really well. I need to make an update video. Thanks again best regards John

[@jamesgarrard4361]

Nice work interesting concept ,i am working on a 6 meter trx 1 watt tx ,si5351a ssb , but i like the fm super work.

[@M0XFXUK]

Hi James, Yes, I figured out how to get FM working properly, I have made a new Si5351 breakout board with an audio input using the Si5351B IC. Keep an eye on the channel as I hope to have a video going through it soon. Thanks for the comment, regards John M0XFX

[@CBRadioWayBack10-4]

in all fairness deviation being around 2.5khz is all you need on a CB/10mtrs.

[@M0XFXUK]

Hi Mark. I was thinking along the same lines, it would have been nice if it could have produced more deviation but it is what it is but it's still a very useful IC considering the price. So I guess we shouldn't lose sight of the fact that it's not really designed for this kind of work and it's a bonus that we can use it in radio applications. I'm still tinkering around with it. 🤪

[@paultomlinson528]

Brilliant work, it sounded ok mate

[@M0XFXUK]

Thanks Paul, I do think It needs a little more work if I'm honest. I feel that the audio quality could be improved by changing the filtering in the audio path at the VC pin, possibly adding an ALC circuit and playing about with the code. I need to spend more time tinkering with it. That being said it's still quite an impressive little IC.

[@paultomlinson528]

@@M0XFXUK I think perhaps a little pre emphasis on higher audio frequencies? But certainly a little tweak... but all thr same well done a brilliant little project and maybe a fancy lcd later... only joking well done mate

[@M0XFXUK]

@@paultomlinson528 Hi Paul yes I agree some higher frequency pre emphasis would most probably improve things. Also I was thinking that an ALC circuit of sorts may improve the overall performance. I am going to keep playing with it when I have time and I will make another video if I can improve the performance. Re the LCD screen and controlling code, I have all ready done this but I am not sure how this would go down on this channel as most people enjoy the repair videos and I think don't really care for C++ programming videos. I am thinking of starting another channel for hobby radio electronics, micro controllers and coding etc but I'm still a little undecided which way to go with it. Thanks for the comment.

[@paultomlinson528]

@@M0XFXUK omg the other channel... sign me up on that one mate

[@rpcomms1]

2-issues I see 1.2995 test set should use 15khz lpf on deviation measurements ,common engineering mistake. 2.You need to audio clip the injected signal. It's needs 20db between 5mV RMS and 35mv RMS max dev limited EU is 1.90khz dev on 1khz test tone 35-50mV input to radio Speech peaks 2.00khz 8ko F3E UK is 2.10khz dev at 1khz test tone into radio 35-50mV RMS Speech peaks 2.20khz dev 11ko F3E By using a simple bat 54 Schottky diode pack,act as a audio diode clipper/limiter after microphone preamplifier (low cut 350hz,hi cut 3.0khz ) Need about 20dB change clean audio About 0.2V audio to drive bat54 diodes,it will clip and square wave the audio,giving it audio punch. On 2995 test set Set FM modulation meter filter to 15khz ,not 0.-3.5khz I'll then see proper clipped full limited FM modulation. I've used this bat54 diode audio clipper on a few CB radios ,inside it's just two diodes back to back. Fed with 0.47-1.0uF caps in and out to prior to dev pot. Very simple trick to get punchy FM audio by audio clipping waveform prior to injected to dev pot,then into your injection point. Hope that helps Rob P RP.Comms

[@M0XFXUK]

Thank you for your informative comment. Yes I totally agree with what you are saying regarding the 2955 setup. I just set the unit up for a rough measurement, as the FM deviation was pretty lousy if I'm honest. I have been tinkering and a lot has happened since I made this video. I have designed and made up a complete PCB with a Si5351B, microcontroller and other circuitry also I have since sorted out the FM modulation by adding pre-emphasis and limiting to the audio signal feeding the VCO input, and it was indeed by adding Schottky diodes and some active filtering using a LM358 op amp. The FM deviation level is now adjustable from around 1Khz to 8Khz. This is a clip of the board in action, and Yes (before anyone mentions) I know playing music into a rig isn't a suitable test. I am planning to make a new video and create a GitHub page as soon as I am back in the UK. Thank you for your input, it's very much appreciated. Check out this link kzbin.info2Pl4KZbrzLM

[@rpcomms1]

@@M0XFXUK Fab news ,well done 👍 I'm very interested in this for a S-NFM,N-Fm+ctcss Analogue hotspot for 24-31mhz band using FM modulation generation this way but I gave up in end was just too complex a few years back and Info was a bit vague. Looks like you cracked it ,well done Lol 15khz lpf on 2995 is classic mistake on FM ,why to they did this never understood and many a radio engineer set up FM deviation wrongly unintentionally 🙄 In workshop lol. Thanks for the link will take a look and finally somebody's cracked it,well done👍 73's Rob

[@M0XFXUK]

@@rpcomms1 Hi Rob, interesting regarding the 2955 deviation 15khz lpf. I will have a look at this at some stage and possibly make a video. Every day is a learning day, as they say, thanks for the heads-up. Re the SI5153B it seems to be working well however; I have not tried CTCSS as yet, but I have a feeling it may just work at the low audio frequencies. It would be a good test and to get CTCSS into the signal. I am more than happy to send you the schematic and an unpopulated PCB as I have a few of them. I am currently in Denmark for work and I should be back in the UK in a couple of weeks time as I plan to make a video and put everything on Github including the code. If you drop me an email, I will ping over the schematic [e-mail] 73s