Frequency division using a JK flip-flop and frequency multiplication using a phase locked loop. assets.nexperia.com/documents/... assets.nexperia.com/documents/...
Thank you, I have been struggling to comprehend the calculations in the datasheet for the vfo freq, your video inspired me to just plug it all together and work it out empirically. I now have a nice stable vfo running at the desired freq. Next the low pass filter, I think this s/b a bit easier.
The 4046 is a great part. You can double frequency with just an XOR if you have a good square wave. You just delay one input with an RC and don't delay the other. Briefly on each edge, the inputs will differ making the output go high. They make 74HC4046 PLLs too. They are faster but most brands are quite buggy. On one I worked with: The Vc pin controls the frequency well in the mid range but up at Vdd, the frequency jumps sky high.
THANKS for the video...I just wanted to comment on the first part where you did a divide by 2 with the flip flop: THE OUTPUT OF THE FLIP FLOP WILL BE 50% DUTY CYCLE, NO MATER WHAT THE 555 DUTY CYCLE IS. PS: What does "JK" stand for?
Very interesting Julian, thank you for this explanation. Do you know what's the best way to divide a 5 GHz frequency so it can be measured by a 100 MHz osciloscope?
Actually, I have seen designs for a frequency multiplier that uses a D flip flop on a PLL. If you're interesting, it's included in the example circuits in Paul Falstad's java circuit simulator, also known as circuitmod for the standalone version.
Thanks Akarsh. I'm still trying to decide how to proceed with the breadboard computer idea. My current thinking is a blinky computer that will only ever blink an LED, but in increasingly sophisticated ways. I'd like the finished design to be an OISC (one instruction set computer).
after watching the video, I got inspired and got my hands on these devices. But even after trying to fine tune VCO to lock onto my external clock (15.5KHz from a function generator), I couldnt get it to lock, it always ends up with jitter in the output. I set my offset at 12KHz.
Nice video, but it would be nice to say what's the purpose of PLL. You could instead use just higher frequency oscillator instead of trying to double the frequency.If you need 2 frequencies is still the same - use higher frequency oscillator and a divider (because it's easier than multiplying).
The 4046 is just about available but how about showing how the same functionality/circuit can be realized in a cpld such as an epm3032. I have yet to play with a cpld due to work commitments (though I have a dev board) but this looks like a good project for everyone to learn how to use a cpld without having to go the whole hog with a fpga which I believe is a different kettle of fish.
Until he got the free scope from Keysight he used a really shit Owon toy scope. I don't think he would ever have bought such an expensive scope as a Keysight, so there's not much point asking what similar scopes he might have considered.
The Owon is actually a nice bit of kit especially when fitted with the internal Li Ion battery. I have a Rigol DS1104Z but the ergonomics are horrible compared with the Owon so the Rigol gets a lot less use. The definition of a good scope is the one given a choice (I have 4) which is automatically reached for first. The Owon in my case.
Nice video! I like a lot reverse engineering and I am trying to find the transmittion protocol between a generic parking sensor module for a car from China and his receiver.The communication is made through 3 wire,Ibelive that is serial communication but I don't have an oscilloscope to identify the sequence.I tried with a CH340 usb to serial converter but I cannot identify the baud rate.The module is from China is around 12$. Please help me if you can,it will be a interesting video to watch not only for me.Thank you.
You'll want a "logic analyzer" something like Saleae Logic. (Some scopes have logic analyzer functions built in to them, and some logic analyzers now have "scope" functions.)
Nevermind the bus pirate or logic analyses. Your sensors are raw 40kHz ultrasonic resonators. Essentially, by themselves they are no different than a high frequency speaker / microphone combo. You have to use support circuitry like in the control unit to use them for distance measurement.
This seems a complicated way of doing the frequency double. The first hit in google gives.. "Differentiate the square wave, full wave rectify and then pass through a Schmidt trigger". The Schmidt trigger is optional. So a resistor, capacitor and four diodes at it's simplest. What I like about this video is how there is no automatic reach for a micro controller. As an exercise it can be very educational to take a circuit problem and then use lower and lower tech ways of implementing a solution. Even using no active devices at all.
To use this approach the input signal must be exactly 50% duty cycle. In that case after frequency doubling, resulting wave is no longer 50% duty cycle, so no further doubling is possible.
So we need the double of the frequency to double the frequency. But if we made the double frequency, then what is the point of the phase locked loop? If we need both frequency, then why not just divide it?
If I understand correctly, we actually need to halve the frequency in order to get double the frequency. But I guess you're referring to where Julian was adjusting the VCO to be around double the frequency? I think the way that works is that if you don't have a frequency counter or oscilloscope, you can "guess" the correct resistor value and if it's close enough, the PLL will still "lock" correctly. I remember having to adjust a PLL for an old computer cassette interface, and the way to do it without a frequency counter (an expensive device in those days) was to put a multimeter across two pins and with a data/program tape playing, adjust the pot until you got the multimeter needle hovering around a certain voltage (I think it was 2.5V but I don't remember for sure). That would set the variable resistance to "close enough". It worked, too.
You take a variable frequency oscillator (VCO), half its output frequency and compare* that to a fixed reference clock (555), negative feedback adjusts the F_VCO until F_VCO/2 = F_CLOCK, then F_VCO must be 2x F_CLOCK. Just like opamp action with voltages. You can do this for much larger divisions than half and many instruments (scopes, spectrum analysers etc) use a 10MHz reference clock to generate hundreds of MHz or GHz through a PLL. *(in the case of PLL, comparing phase, but you can compare just frequency in a FLL, the difference being the 2 clocks won't necessarily be in phase with each other)
It was a bit unclear. If you have a flip-flop in the loop dividing by two, the PLL output will be double the input frequency. But if you used a counter to divide by 3, 4, or 5, the output would be 3x, 4x, or 5x the input frequency.
Precision! The PLL oscillator doesn't need to be precise. You can get really high frequencies out of an RC oscillator, but precision is maybe 10%. Quartz crystals are super-precise in their frequency, but they can't go above 300MHz. So your computer, for example, uses a PLL to generate the multi-gigahertz clock frequency while keeping it in phase with a precise but much lower frequency quartz oscillator. They are also used when a circuit needs to operate in sync with an external clock but at higher frequency, because several internal steps are pipelined.
Julian, appropos your series on the Z80 (8086), I have been pulling some chips from some old equipment and have an NEC (Ireland) SN8255ac-2 Programmable Peripheral Interface going spare if you want it? Datasheet here :- pdf.datasheetcatalog.com/datasheet_pdf/nec/UPD8255AC-2_and_UPD8255AC-5.pdf If you want it for a future video just drop me a message and I will post it to you.
Much much cheaper. At the expense of any country whose population is generally complacent with no patience to the point where nothing will stop the purchase from China.
Ahhh ok I thought it might have been one of those electronic kits with all the chips leds resistors etc already onboard and we just interconnect them. Thanks for reply
I liked your videos, even if you had a cheap Owon oscilloscope. But since you got a very expensive oscilloscope as a gift, and you always brag with it, you lost my sympathy. If you have such a good oscilloscope at least don't humble down it with 100 kHz projects. Your old Owon is also very good for such projects. Do not brag! I am sure that noone from your watchers will buy it in the future.
