Do NOT use MPF-102 in the ASM-1!

[Magnus Danielson]

>   Hi Guys -
> 
>   I must humbly issue a recall on my statement last week that the Radio
>   Shack MPF-102 can be used as a substitute for the 2N???? / NTE466 FET in
>   the ASM-1 VCO.
> 
>   The MPF-102 is a FET and does have the same pinout, but it is not the
>   same.
> 
>   After building a VCO, I was having a problem with the output level of the
>   sawtooth and didn't think it had anything to do with the FET. I mean it
>   was working and tracking and making sawtooth waves. I put it down for a
>   few days, came back and tried it and it wasn't working at all. Testing
>   showed weird results and the 3140 was dead. I replaced the 3140 and -
>   just to be sure, replaced the MPF-102 with an NTE466. This worked great
>   first try and has worked fine ever since.
> 
>   I'm really sorry if anyone followed my advise and got stuck -
>   Get a NTE466 or something else as a replacement.

I had a similar experience with the 2N4391's that I am using.

Did your 3140 "hang" at +15V?

Mine did and the LM311 kept ticking at HF (some 100 kHz to 500 kHz...)...

I hooked an 30 pF cap from the input to ground... this slowed things down and
allowed the FET to have time enougth to empty to cap... can something similar
maybe work for the MPF-102's?

Changing the 18 pF will also do the trick... but I never bothered...

Oh, did you know that the VCO is by design (quite) insensitive to power 
voltage?

Another thing, the Linear FM inputs are negative, an increased input voltage
will decreese the frequency...

The power supplies can however change the output level of the sawtooth 
amplitude. One can stabilize it with a voltage reference with a few simple
steps... but I guess few people are that picky....

There is a limit in how deep FM migth be done since the circuit can't make the
capacitor to discharge... but maybe that is not a big problem to most people.

The VCO with the given components have a nonmodulated frequency at about A5
(which is 880 Hz) or rather 910 Hz. Inserting a trimmer either at the 1.5 Mohm
resistor or better at the Lin FM input allows for preciser trimming of the
nonmodulated frequency. After the nonmodulated frequency has been trimmed will
the scale and initial frequency be oversimple to trim, "loop" trimming becomes
unessecary or shorter.

Cheers,
Magnus