Hello Neil

Thanks for the hint. I did some further experiments the last hour, and
can confirm, that the CP1 is more stable on a only slightly varying
master frequency, but it would not cover a wider frequency range (more
than an octave) of the input signal.

For varying frequencies I get the best results with CP3. In the range of
the input signal between 80 and 400Hz I get up to two and a half octaves
- but for the Din-Sync clock multiplication I will have input signals
between 10 Hz and 100Hz. I tried with various filters (esp. the
slewlimiter), but I do not get a satisfying range with this low input
frequency.
I can make it track fine, if I send an additional control voltage to the
PLL's VCO and to the VC-slewlimiter, that corresponds to the frequency
of the input signal. But this will not be available in the case of the
DIN-Sync multiplication.

I think for my DIY project I will work with a two stage approach.
Part 1 will be counter based F/V converter, which provides a rough
control voltage.
Part 2 will be the PLL that does the fine tuning.

Nevertheless if anyone can provide reliable settings for low frequency
syncing with an wide range, any help is appreciated.

> you are looking to do a x40 multiplication which using the A160/161

is not going to work

> (sorry you clearly know this already),

Yes ;-)
It won't work only the 160/161 or with a standard 160/161 . But there
are several approaches to achieve it:
The first is: use logic gates, that read decimal 41 = binary 101001 from
the output of 160 and send the output to the reset.
The second is: take the 1/8 signal from the A-160 and trigger with it
the A-161 and use the step 6 as reset. This requires a modified A-161
that has been extended with a separate clock input.

But in fact, I want to use the A-100 system experiment only with the
PLL. Therefor it is not necessary to have the true division factor. The
final device will be a custommade standalone unit, where the divider is
made from a CD4520 and two ANDs.

Florian

Am 23.04.2015 um 08:56 schrieb Neil Kagan blinkenlicht@...
[Doepfer_a100]:

> I don't know if there is an 'ideal' setup. Certainly CP1 seems more
> stable than the other two, as it locks at harmonics. I think you might
> find that it is going to be a process of just trying things until it
> works... One thing I did notice is that you are looking to do a x40
> multiplication which using the A160/161 is not going to work (sorry
> you clearly know this already), so would the VC frequency divider be a
> better option