[sdiy] Really Low, LFO

[Tom Wiltshire]

My PIC LFO goes down to 0.05Hz (20 seconds) as it stands. If you were  
to slash a factor of 256 out of the sample rate (to a 76Hz sample  
rate!), that'd go down to 0.000195Hz, 85 mins a cycle. I agree with  
what others have said, it isn't hard to do *low* frequencies with  
digital. It's when you go *up* that things get difficult!

T.



On 2 Apr 2010, at 22:45, <jays at aracnet.com> wrote:

> I think I'd agree here, I'd go digital. I do think a DSP would be  
> overkill. Especially with what you get in low end uCs like PICs now  
> days.
>
> When you have timing this long the quality of parts becomes  
> critical. Things like op-amp offsets start to affect symmetry. I  
> now use LT1013s in some of my LFOs because of this. As mentioned  
> issues with caps become critical and so on.....
>
> Jay S.
>
> cheater cheater wrote:
>
>> Leakage current of your capacitor and off-resistance of the  
>> transistor
>> used for resetting the core are going to be the issues, I think.
>>
>> If you want an infrasonic LFO I suggest doing it in DSP, there's no
>> advantage to using analog at those frequencies that I can think of.
>>
>>
>> On Fri, Apr 2, 2010 at 22:51, Kyle Stephens <lightburnx at yahoo.com>  
>> wrote:
>>> Can anyone point me to some reading material on infrasonic  
>>> oscillator design? I want to build a really low, low LFO, with a  
>>> period of around half an hour (yes, _that_ slow!).
>>>
>>> As HF designs have their share of issues that get worse in that  
>>> spectrum, infrasonic has its problems too I would imagine - I  
>>> just don't know what exactly they might be.
>>>
>>> For my app, accuracy isn't as important as precision/ 
>>> repeatability, though if both are possible then both are welcome.
>>>
>>> Closest I've found so far is this:
>>>
>>> http://tinyurl.com/yf9wzp3
>>>
>>> I reckon I can tailor it if I make the integrator cap several  
>>> decades larger, to shift that 0.1Hz to 100KHz range several  
>>> decimal places downward.
>>
>> Eh, not really about the size of the cap, it's more about being able
>> to control the current accurately enough.
>>
>>> I've a copy of AD's Nonlinear Circuits Handbook, which has some  
>>> material on the topic, though nothing too specific on freqs  
>>> beneath 1 Hz (half an hour is ~ 0.0001 Hz).
>>>
>>> I spoke to Prof. Lanterman about this a while back, and he  
>>> recommended a microcontroller based solution,
>>
>> Yep, I second that. DSP options (and for an lfo you don't even need a
>> dedicated dsp uC such as a 56k) give you accuracy AND repeatability;
>> but they're easy  to badly implement on their own (aliasing, ...)
>> although, comparing the typical sampling frequency you'd run this at
>> to the frequency you expect, the error in pitch coming from aliasing
>> on the discontinuity (which gives you only an error of +-1 sample,
>> maybe 2) will be minuscule. Your biggest problem will be implementing
>> an accumulator register big enough, I think.
>>
>> D.
>>
>>> and while that's something I want to break into, it's a whole  
>>> 'nother can of worms too. If this can be done in analog, it'd be  
>>> a cool achievement and an enrichment exercise if nothing else... :]
>>>
>>>
>>>
>>> _Kyle (so slow it Herz?)
>>>
>>>
>>>
>>> _______________________________________________
>>> Synth-diy mailing list
>>> Synth-diy at dropmix.xs4all.nl
>>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>>>
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at dropmix.xs4all.nl
>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy