MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

LM565 VCO ?
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author LM565 VCO ?
Yekuku
Hi wigglers,
I was given a few LM565 Phase Locked Loop ICs by a friend, and I thought that it would be interesting if its VCO frequency could be exponentially controlled.
I searched the internet but i could not find any examples of using the 565 as a VCO in synths.
Did anybody try experimenting with it ? Any pros or cons ?

I have already built Thomas Henry 4046 VCO, and I am hopping that i can get similar results from the 565. ( lol! high hopes !)
I am a complete noob in engineering /designing circuits , I am just starting out and experimenting in order to learn a few new things. So far I have studied Thomas Henry designs via Scott Stites site , and I own a big THANK YOU to Mr Henry and Mr Stites for sharing their wisdom and helping all of us.
I have breadboarded the LM565 and with minimal components , I have managed to linearly control the frequency from 25Hz to 12kHz and get a triangle waveform from pin 9 (timing capacitor pin).
One problem i have is that in low frequencies (<100Hz), the triangle wave looks like a shark fin, but as I increase the frequency the symmetry of the triangle wave is getting normal. I am wondering if this issue could be solved...
I am looking at Mr. Henry's classic exponential converter and thinking of connecting to pin 8 (timing resistor pin) , do you think that i will have any luck controlling the VCO exponentially?


Any ideas/ help / guidance would be deeply appreciated smile
dot matrix madness
I guess the most obvious way is to check if the 565 can be controlled in the same way as in the 4046 in TH's design, which you mentioned. Which means first comparing the data sheets/pin outs of both chips.
Yekuku
Both chips are PLLs but unfortunately they dont have the same functions.

here is the 565 pinout:


here is the 4046:


Mr Henry did some of his magic to get the triangle waveform out of the 4046 and I highly doubt that the same technique can be applied to the 565.
At the time being I am only using the ICs timing capacitor pin 9 to get the triangle without any magic involved, thats why I probably get the "shark fin" problem that I have mentioned earlier. Hopefully using Mr Henry techniques, forming sine and pulse waves from the triangle should not be problem.
So the question is : will it track right ?
For the exponential control I am looking at Mr Henry's exponential current mirror technique.
I would love to hear other people experiences with the 565, do you think that it could be used for musical applications?
guest
run the LM565 between 0V and -12V. connect VCO CV to 0V. make a PNP exponential converter, and drive the current into the timing resistor input.

there isnt much to be done about the shark fin curve. the timing capacitor is internally buffered with an NPN transistor, which has some base current. this degrades performance at the low end. you can run at higher currents, but this will degrade performance at the higher end.
Yekuku
thank you for you help Guest , I deeply appreciate it.

So far all I have done is presented in this lousy schematic :


Using a single line power supply to provide 15V, I have managed to linearly control the frequency from 25hz to 12 Khz with a 100k potentiometer.
Please note that I have not connected anything to Pin7 (Vco control voltage), I am just using the resistor and capacitor timing pins ( 8 & 9) to get a wide frequency range.

I would like to ask you some questions in order to clarify a few things.
Quote:
run the LM565 between 0V and -12V. connect VCO CV to 0V. make a PNP exponential converter, and drive the current into the timing resistor input.

Why should I use reversed polarity ( 0v for Vcc+ and -12V for Vcc-)?
Why should I tie the VCO control voltage to 0V?

Quote:
there isnt much to be done about the shark fin curve. the timing capacitor is internally buffered with an NPN transistor, which has some base current. this degrades performance at the low end. you can run at higher currents, but this will degrade performance at the higher end.

thanks for the logical explanation, I will try higher currents and see if things can get better.

Please bear with me, as said I am noob at this , I dont want to build just another VCO, all I want is to experiment and learn new things.
guest
if you want to control the frequency exponentially (you can already do it linearly with the CV input), then you need to feed it an exponential current into the timing resistor input (pin8 on your schematic). there are 2 ways of doing this. one is to use a standard NPN exponential converter (which sinks current from a voltage higher than 0V), and then put a current mirror up at +Vcc that reflects the current back down into pin8. the other way is to use a PNP exponential converter, which sources current into voltages less than 0V.

if you go with the PNP approach (which is a bit more accurate and fewer parts), the CV pin needs to be 0V or less (0V is probably the best setting). you can run the LM565 at +/-12V, but its less current draw to just run it at 0V/-12V (or -15V in your case).

the voltage at the timing resistor pin is determined by the CV voltage. so if you keep this fixed, then there are fewer variations on your exponential current source, and it is more accurate.

when you meausred the triangle output, did you buffer with an opamp before looking at it on a scope?
Yekuku
thank you for the explanations and all the valuable information , it is quite clear now.
I had no idea that I could use PNP transistors for the exponential converter, so far I have only used NPNs (3904).
I am thinking of using a pair of 3906s for the PNP exponential converter, should the transistors be matched in order to have success in exponential tracking, or the matching affects only temperature drifts ?


Quote:
when you measured the triangle output, did you buffer with an opamp before looking at it on a scope?

No, I did not used a buffer , I observed the waveform from pin 9.
Do you think that the oscilloscope probe is loading down the triangle output, thats why I get a shark fin in lower frequencies ?
Also I need to take care of the waveform's offset, I did not make up my mind yet if I will use just a decoupling capacitor or a differential amplifier.
At the time being i am focused on the exponential control, I am searching for PNP exponential converters right now, hopefully in a few days I ll get back with my findings.
many thanks !
guest
to use PNP transisstors, its the same circuit as NPN, but the resistor at the collector of the reference transistor goes to -Vcc rather than +Vcc. transistor matching is mostly for temperature compensation, so you can definitely get something up and running without matching, but it will probably drift a fair bit.

i would reccomend a buffer at the output, that is definitely going to cause the sharkfin if you do not have it. you can adjust offset with that opamp as well.
Grumble
Besides that, it is good to know that the higher the frequency the harder it is to hear the difference in wave form.

MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group