Doepfer

Dear list members,

I was under the impression that a high pass filter should let a 
signal pass almost unchanged with the cut-off frequency all the 
way down (the opposite of a low pass). However my A-121 
seems to be cutting higher frequencies at low cut-off settings 
(resonance all the way down), more like a band pass filter.
The result is especially clear when runnig a rich external signal 
through it, songs on the radio in my case.

When viewing an A-121's sine wave output on an oscilloscope, 
the "Band" output seems to be the only output to generate a true 
sine wave. The waveforms from "High" and "Low" seem to be 
clipped at the highest and lowest points, respectively, and the 
"Notch" output is significantly lower than the others. 
Is this an anomaly of my module or do all A-121's behave like 
this?


And which of the 4 lengths of cables available for the A-100 are 
the white ones?

Thanks
M. Dimm

Hi M. Dimm

My A-121 behaves pretty much as you describe it: all the little odd
things you mention are almost certainly due to the CEM3320 VCF chip
that does all the work. The datasheet for the 3320 mentions (for
self-oscillation) that 'the sinewave is not clipped by the first
stage', so this would seem to equate to the  bandpass output in the
A-121 configuration - if one wants to use the filter as an oscillator,
then this is clearly the best place to take the output from (unless of
course the distortion is desired!).

The problem of the highpass output (with a low cut-off frequency)
acting more like a bandpass filter, whilst annoying, is also likely to
be a limitation within the chip. The chip contains 4 'gain cells'
which act in a similar manner to VCAs, and are connected together with
various resistors and capacitors to give the filtering action. The
gain setting is used to control the filter cut-off frequency, and the
gain they can provide probably decreases as the input frequency
increases. Thus for a low cut-off frequency, their gain is already set
quite low: passing high frequencies through them will decrease the
gain further, and thus higher frequencies get attenuated. Standard op
amps (the staple component in most modules) also behave like this, and
in fact the 'emphasis' filter in the A-199 reverb module appears to
make good use of this effect - it is really a high pass filter, turned
into a bandpass one by the falling gain available from the op amp as
the frequency increases. I don't have an A-123 high pass module, but I
note that it uses the same CEM 3320 chip - it would be interesting to
hear from other list members whether it behaves in a similar fashion.

This probably isn't much consolation to you, merely confirming that
you are stuck with this behaviour! (Can't help with the white patch
cord length!)

Tim

[The views expressed above are entirely those of the writer and do not
represent the views, policy or understanding of any other person or
official body.]

> I was under the impression that a high pass filter should let a 
> signal pass almost unchanged with the cut-off frequency all the 
> way down (the opposite of a low pass). However my A-121 
> seems to be cutting higher frequencies at low cut-off settings 
> (resonance all the way down), more like a band pass filter.
> The result is especially clear when runnig a rich external signal 
> through it, songs on the radio in my case.
> 
> When viewing an A-121's sine wave output on an oscilloscope, 
> the "Band" output seems to be the only output to generate a true 
> sine wave. The waveforms from "High" and "Low" seem to be 
> clipped at the highest and lowest points, respectively, and the 
> "Notch" output is significantly lower than the others. 
> Is this an anomaly of my module or do all A-121's behave like 
> this?
> 
> 
> And which of the 4 lengths of cables available for the A-100 are 
> the white ones?
> 
> Thanks
> M. Dimm

Hi Tim

There is a modification that gives a much better highpass behaviour at low
cutoff frequencies to the A-121. I don't know whether it is documented in
the service manual. But this may cause a uncontrolled and unwanted self
oscillation in high cutoff frequencies.

> I don't have an A-123 high pass module, but I
> note that it uses the same CEM 3320 chip - it would be interesting to
> hear from other list members whether it behaves in a similar fashion.
No. As far I understood the problem appears only in the state-variable
circuit. And the 123 is a pute pure hipass filter.

Florian
-- 
Florian Anwander                  |ConSol*  HP-Support
Tel.   +49.89.45841-133           |Consulting&Solutions Software GmbH
Fax    +49.89.45841-139           |Franziskanerstr. 38, D-81669 München
email: florian.anwander@consol.de |http://www.consol.de

Hi Florian,

> There is a modification that gives a much better highpass behaviour
at low
> cutoff frequencies to the A-121. I don't know whether it is
documented in
> the service manual. But this may cause a uncontrolled and unwanted self
> oscillation in high cutoff frequencies.

Thanks, that's interesting to know. The service manual potentially
contains much useful info, but the price rather puts me off!!

> > I don't have an A-123 high pass module, but I
> > note that it uses the same CEM 3320 chip - it would be interesting to
> > hear from other list members whether it behaves in a similar fashion.
> No. As far I understood the problem appears only in the state-variable
> circuit. And the 123 is a pute pure hipass filter.

This is also nice to know, thanks again!

Tim

[The views expressed above are entirely those of the writer and do not
represent the views, policy or understanding of any other person or
official body.]

hi,
you mean the grey cables lenght??
they are 30cm.there are no white cables from doepfer.
bakis.

--- stinchcombe_t <tstinchcombe@qinetiq.com> wrote:
> Hi M. Dimm
> 
> My A-121 behaves pretty much as you describe it: all
> the little odd
> things you mention are almost certainly due to the
> CEM3320 VCF chip
> that does all the work. The datasheet for the 3320
> mentions (for
> self-oscillation) that 'the sinewave is not clipped
> by the first
> stage', so this would seem to equate to the 
> bandpass output in the
> A-121 configuration - if one wants to use the filter
> as an oscillator,
> then this is clearly the best place to take the
> output from (unless of
> course the distortion is desired!).
> 
> The problem of the highpass output (with a low
> cut-off frequency)
> acting more like a bandpass filter, whilst annoying,
> is also likely to
> be a limitation within the chip. The chip contains 4
> 'gain cells'
> which act in a similar manner to VCAs, and are
> connected together with
> various resistors and capacitors to give the
> filtering action. The
> gain setting is used to control the filter cut-off
> frequency, and the
> gain they can provide probably decreases as the
> input frequency
> increases. Thus for a low cut-off frequency, their
> gain is already set
> quite low: passing high frequencies through them
> will decrease the
> gain further, and thus higher frequencies get
> attenuated. Standard op
> amps (the staple component in most modules) also
> behave like this, and
> in fact the 'emphasis' filter in the A-199 reverb
> module appears to
> make good use of this effect - it is really a high
> pass filter, turned
> into a bandpass one by the falling gain available
> from the op amp as
> the frequency increases. I don't have an A-123 high
> pass module, but I
> note that it uses the same CEM 3320 chip - it would
> be interesting to
> hear from other list members whether it behaves in a
> similar fashion.
> 
> This probably isn't much consolation to you, merely
> confirming that
> you are stuck with this behaviour! (Can't help with
> the white patch
> cord length!)
> 
> Tim
> 
> [The views expressed above are entirely those of the
> writer and do not
> represent the views, policy or understanding of any
> other person or
> official body.]
> 
> 
> > I was under the impression that a high pass filter
> should let a 
> > signal pass almost unchanged with the cut-off
> frequency all the 
> > way down (the opposite of a low pass). However my
> A-121 
> > seems to be cutting higher frequencies at low
> cut-off settings 
> > (resonance all the way down), more like a band
> pass filter.
> > The result is especially clear when runnig a rich
> external signal 
> > through it, songs on the radio in my case.
> > 
> > When viewing an A-121's sine wave output on an
> oscilloscope, 
> > the "Band" output seems to be the only output to
> generate a true 
> > sine wave. The waveforms from "High" and "Low"
> seem to be 
> > clipped at the highest and lowest points,
> respectively, and the 
> > "Notch" output is significantly lower than the
> others. 
> > Is this an anomaly of my module or do all A-121's
> behave like 
> > this?
> > 
> > 
> > And which of the 4 lengths of cables available for
> the A-100 are 
> > the white ones?
> > 
> > Thanks
> > M. Dimm
> 
> 


=====
synthfreak(parallel worlds)
athens-greece
[Doepfer_a100] group owner-moderator

__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://taxes.yahoo.com/

oops!!!!!wrong....the grey cables are 50cm long.
sorry,
bakis.


--- stinchcombe_t <tstinchcombe@qinetiq.com> wrote:
> Hi M. Dimm
> 
> My A-121 behaves pretty much as you describe it: all
> the little odd
> things you mention are almost certainly due to the
> CEM3320 VCF chip
> that does all the work. The datasheet for the 3320
> mentions (for
> self-oscillation) that 'the sinewave is not clipped
> by the first
> stage', so this would seem to equate to the 
> bandpass output in the
> A-121 configuration - if one wants to use the filter
> as an oscillator,
> then this is clearly the best place to take the
> output from (unless of
> course the distortion is desired!).
> 
> The problem of the highpass output (with a low
> cut-off frequency)
> acting more like a bandpass filter, whilst annoying,
> is also likely to
> be a limitation within the chip. The chip contains 4
> 'gain cells'
> which act in a similar manner to VCAs, and are
> connected together with
> various resistors and capacitors to give the
> filtering action. The
> gain setting is used to control the filter cut-off
> frequency, and the
> gain they can provide probably decreases as the
> input frequency
> increases. Thus for a low cut-off frequency, their
> gain is already set
> quite low: passing high frequencies through them
> will decrease the
> gain further, and thus higher frequencies get
> attenuated. Standard op
> amps (the staple component in most modules) also
> behave like this, and
> in fact the 'emphasis' filter in the A-199 reverb
> module appears to
> make good use of this effect - it is really a high
> pass filter, turned
> into a bandpass one by the falling gain available
> from the op amp as
> the frequency increases. I don't have an A-123 high
> pass module, but I
> note that it uses the same CEM 3320 chip - it would
> be interesting to
> hear from other list members whether it behaves in a
> similar fashion.
> 
> This probably isn't much consolation to you, merely
> confirming that
> you are stuck with this behaviour! (Can't help with
> the white patch
> cord length!)
> 
> Tim
> 
> [The views expressed above are entirely those of the
> writer and do not
> represent the views, policy or understanding of any
> other person or
> official body.]
> 
> 
> > I was under the impression that a high pass filter
> should let a 
> > signal pass almost unchanged with the cut-off
> frequency all the 
> > way down (the opposite of a low pass). However my
> A-121 
> > seems to be cutting higher frequencies at low
> cut-off settings 
> > (resonance all the way down), more like a band
> pass filter.
> > The result is especially clear when runnig a rich
> external signal 
> > through it, songs on the radio in my case.
> > 
> > When viewing an A-121's sine wave output on an
> oscilloscope, 
> > the "Band" output seems to be the only output to
> generate a true 
> > sine wave. The waveforms from "High" and "Low"
> seem to be 
> > clipped at the highest and lowest points,
> respectively, and the 
> > "Notch" output is significantly lower than the
> others. 
> > Is this an anomaly of my module or do all A-121's
> behave like 
> > this?
> > 
> > 
> > And which of the 4 lengths of cables available for
> the A-100 are 
> > the white ones?
> > 
> > Thanks
> > M. Dimm
> 
> 


=====
synthfreak(parallel worlds)
athens-greece
[Doepfer_a100] group owner-moderator

__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://taxes.yahoo.com/

--- In Doepfer_a100@y..., Florian Anwander <
Florian.Anwander@c...> wrote:
> Hi Tim
> 
> There is a modification that gives a much better highpass 
behaviour at low
> cutoff frequencies to the A-121. I don't know whether it is 
documented in
> the service manual. But this may cause a uncontrolled and 
unwanted self
> oscillation in high cutoff frequencies.
> 
> > I don't have an A-123 high pass module, but I
> > note that it uses the same CEM 3320 chip - it would be 
interesting to
> > hear from other list members whether it behaves in a similar 
fashion.
> No. As far I understood the problem appears only in the state-
variable
> circuit. And the 123 is a pute pure hipass filter.
> 
> Florian

That was going to be my next question, whether the A-123 has 
the same characteristics. 

I'd be interested to know more about that modification to the A-
121 though.

Thanks to all for the extensive information.

M. Dimm.

Dear mdimmm 

> I'd be interested to know more about that modification to the A-121 though.
I'll have a look at the Service manual at home

Florian
-- 
Florian Anwander                  |ConSol*  HP-Support
Tel.   +49.89.45841-133           |Consulting&Solutions Software GmbH
Fax    +49.89.45841-139           |Franziskanerstr. 38, D-81669 München
email: florian.anwander@consol.de |http://www.consol.de

Hi M Dimm

> > > I don't have an A-123 high pass module, but I
> > > note that it uses the same CEM 3320 chip - it would be 
> interesting to
> > > hear from other list members whether it behaves in a similar 
> fashion.
> > No. As far I understood the problem appears only in the state-
> variable
> > circuit. And the 123 is a pute pure hipass filter.
> 
> That was going to be my next question, whether the A-123 has 
> the same characteristics.

The above is no longer true! I treated myself to some new modules last
week, including the A-123.  My initial reaction is that the A-123
behaves in pretty much the same fashion as the A-121, and in fact
could be slightly worse. I haven't really had time to check it out
properly though, I'll try and do so over the weekend, and then I'll
report back!

Tim

[The views expressed above are entirely those of the writer and do not
represent the views, policy or understanding of any other person or
official body.]

Hi Tim


> The above is no longer true! I treated myself to some new modules last
> week, including the A-123.  My initial reaction is that the A-123
> behaves in pretty much the same fashion as the A-121, and in fact
> could be slightly worse. I haven't really had time to check it out
> properly though, I'll try and do so over the weekend, and then I'll
> report back!
Uuups, I just remember, that my A123 is one of those modules with a
preseries frontplate, but should be electronically the same as the series
modules (some parts soldered on the other side). I cannot confirm that it
is REALLY a electronically identical. Also I have to look.

Florian
-- 
Florian Anwander                  |ConSol*  HP-Support
Tel.   +49.89.45841-133           |Consulting&Solutions Software GmbH
Fax    +49.89.45841-139           |Franziskanerstr. 38, D-81669 München
email: florian.anwander@consol.de |http://www.consol.de

> week, including the A-123.  My initial reaction is that the A-123
> behaves in pretty much the same fashion as the A-121, and in fact
> could be slightly worse. I haven't really had time to check it out

I can happily report that the A-123 is in fact *considerably better* 
than the A-121 in this respect! (For my initial 'look-see'  I used an 
A-110, but since the sine wave is quite 'approximate', it gets badly 
mangled by the filter, which tends to obscure things a bit! The sine 
output of my A-111 is clearly much closer to a pure tone, and thus 
comes through the filter less mangled, so it's easier to see what is 
going on.) By setting the VCO to a specific frequency and monitoring 
the voltage levels out of the filters, I adjusted the frequency 
setting on the filter so that the given frequency was the -3dB lower 
cut-off frequency, then I increased the VCO frequency until I went 
through the passband and came to the upper -3dB point. (I had 
resonance set to minimum throughout.) In all cases I tried, regardless 
of the Freq knob setting, the A-123 upper frequency is at about 30kHz: 
for the A-121 it was as follows:

Lower cut-off freq = 25Hz (approx F=2.5 on 121, F=2 on 123), upper 
cut-off = 800Hz
lower = 50Hz (F=3.5 &F=3), upper = 1.6kHz
lower = 100Hz (F=4.5 & F=3.5), upper = 3.3kHz
lower = 200Hz (F=5 &F=4.5), upper = 6.6kHz
lower = 400Hz (F=5.5 & F=5), upper = 11.5kHz

Setting the A-121 to give the upper cut-off at 15kHz, gives the lower 
as 600Hz, at F=6. At low F settings, 0 - 2, on the A-121 there comes a 
point where you don't really get a well defined passband at all! This 
clearly supports M Dimm's findings, i.e. that the A-121 high pass is 
really more bandpass, especially at low F knob settings!

The intro paragraph to the high pass chapter of Don Lancaster's 
'Active Filter Cookbook' puts this problem rather bluntly: "In 
reality, there is no such thing as an active high-pass filter, for the 
upper-frequency rolloff of the operational amplifiers in use will 
combine to give a passband. If an active filter is to be good for 
anything, we have to save enough 'daylight' between the lower passband 
limit of the active circuit and the upper limit set by the op amp." 
The A-123 obviously achieves this, the A-121 less so. Haven't managed 
to fathom out why the A-121 configuration should be so much worse, and 
unfortunately there is unlikely to be enough info on the CEM datasheet 
to be able to work it out, so its likely to remain a mystery to me 
(unless Dieter is reading and can offer some explanation....)

Tim

[The views expressed above are entirely those of the writer and do not 
represent the views, policy or understanding of any other person or 
official body.]