Neutron Filter Build Report

[primalphunk]

I have no real previous electronics background but I felt gutsy after doing about 7 other easier projects so this is what I tackled.  Considering my almost total ineptitude I am pleasantly surprised that this project is working at all.  This is a testament to just how easy a hard project can be made thanks to RG's cool Mutron III workalike, the PCBs that JD sells at GGG, Aron's cool forum, the parts that Steve sells at Small Bear and of course many great pieces of advice provided by Mark Hammer on the topic of envelope filters.  

Right now my unit is not sounding that great but I anticpated the problems I would encounter before finishing thanks to this forum.  On my first attempt to use the unit I discovered that it was doing almost nothing to my signal.  After spending entirely too much time trying to track down the problem I realized that the unit was just not turning on properly.  I'm building this as part of a half rack unit that houses 4 projects in total so I'm not using batteries for power or stereo jacks for switching.  Since I had already populated the board with the switching transistor and all the related charge pump components I just needed to add a jumper to ground at the stereo lug input.  Woooohooo!  Now it works...just not very nicely yet.  

I used the H11F3 and really don't like all the distortion I am getting so I have now ordered the Hamamatsu photo coupler from Small Bear.  A potential problem I was able to anticipate from this forum.  Eventually I noticed some whining coming from my unit.  No worries, this to was anticipated thanks to this forum.  So I'm gonna get the 1044 charge pump that seems to work better for this project.  Guess what?  This problem was also anticipated thanks to the forum.

I'll post again after the last two items get switched out and I'm expecting that I'm gonna like the results.

 

[Joe Viau]

How did you make out? I was planning on building a Neutron with the H11F3.  Do the other optoisolators work better?

[Bill Bergman]

I've built it both ways and preferred the H11F3's

[Transmogrifox]

I have had good results making my own LDR/LED assemblies.  I buy the assorted bunch from Rat Shack and sit around with my DMM next to a bright lamp and measure them to find a pair that come pretty close to eachother.

After this, I epoxy them together and paint them 5 or 6 coats with black fingernail polish (essentially until light in the room stops effecting them much).

I do some further testing with the DMM and a battery and Pot to determine the currents for maximum and minimum resistance that I want to see (since the CDS cells I select range from about 4Meg full dark, to 10k full bright).

Anyway, much of the distorted sound is from the envelope follower.  The Mutron/Neutron EF section is still pretty turbulent on the output to the LED, so some LDRs with fast responses will feed some of this rectified trash through into the audio path.

The envelope filter I designed was similar to the Neutron, except it uses a 5th order lowpass filter on the envelope follower to take audio range trash out of the response.

Unfortunately, the project took me 3 months to complete, summing to nearly 150 hours of designing, building, experimenting and troubleshooting--and yes, I added a lot of options and pots for parameters that the Mutron does not offer that added much time and complexity to the design.

The good part is that it's a fun toy to play with now.

[kerryandray]

I have had good results making my own LDR/LED assemblies.  I buy the assorted bunch from Rat Shack and sit around with my DMM next to a bright lamp and measure them to find a pair that come pretty close to eachother.

After this, I epoxy them together and paint them 5 or 6 coats with black fingernail polish (essentially until light in the room stops effecting them much).

I do some further testing with the DMM and a battery and Pot to determine the currents for maximum and minimum resistance that I want to see (since the CDS cells I select range from about 4Meg full dark, to 10k full bright

Hey, Transmogrifox...I just built myself an LED/LDR assembly (not quite finished...I still need to apply something black to the ends). 

I am trying to understand what you do when you "do some further testing with the DMM and pot".  What specifically are you doing, how do you use your readings to determine "the currents for maximum and minimum resistance that I want to see", and what do you do once you know the currents you want to see?  Is it to set the value of Resistor X in the neutron filter schematic?

As a reference, My LED is a red diffused with a forward voltage between 1.9 and 2.1.  My LDRs range from about 4.6M and 900 ohms.

Thanks in advance for any help you can offer, it would be greatly appreciated!

Ray

[Transmogrifox]

Referring to this:
http://www.geofex.com/PCB_layouts/Layouts/neutronpub.pdf

Really I think that article gives all the information needed to successfully build the Neutron with any vactrol with adequate min and max resistance ranges.

Here's some more in case you find it hard to digest (maybe this will only confuse you more, but here I will try  )
Temporarily replace Rx with a pot and adjust it until you have the right amount of range. When you get the sweep sounding right then take the pot out of the circuit, measure it, and put in the nearest standard resistor value to what you measured.

Actually RG gives some instruction in the text about experimenting with fixed resistor values.  This works the same as long as you have a resistor kit with a lot of different values to try.

One extra subtle adjustment that RG does not include in the Neutron schematic is a resistor at the junction of Rx-LED-R17 and to -9V.  This would probably be in the 50k to 1 Meg range (a 1Meg pot would be good for experimenting).  You can use this to adjust the "take-off" of the sweep.  This pre-biases the LED with a steady state curent.

The thinking is that from something like 4.7 Meg to around 500k the filter does not move much.  On sweep attack it starts up a little more abruptly when the LED starts dark. 

The extra resistor can be adjusted to start it at a "dark" resistance that isn't 100% dark and the pick onset doesn't feel "gated" as much.  Anyway you would adjust the pot until the filter settles to a minimum "depth" that you like.

Both of these adjustments could be added as regular controls where you use a pot instead of a fixed resistor and make it a tweakable knob on the top of the box.  Rx adjusts "sensitivity" or range while the resistor between Rx and -9V serves as "depth".  You can make this parameter completely controllable to the extreme limits of the CdS cell resistance by removing R8 and R9, or changing them to a larger value if you want to limit it somewhere.

When I was doing "experiments with the DMM and pot" I was basically just tracing the LED current vs LDR resistance.  I hooked up a battery in series with a pot wired as a rheostat (Leg A -- Wiper, and Leg B disconnected) and another series resistor.  I would adjust the pot, measure voltage on the series resistor and use Ohm's law to get the LED current.  Then I would measure the LDR resistance and enter the LED current and LDR resistance into a column in my notebook.  After enough of these measurements I had enough points to draw a decent enough graph of LED current vs LDR resistance.

I used this to design the rest of the circuit around it.  In the case of the Neutron, you already have the circuit so you just need to adjust the max current to the LED to suit what the circuit expects. Optionally you can add another resistor to adjust the min current in the LED to change the pick response in a subtle way.  Don't do that unless you do the adjustment of Rx and find you're not completely satisfied with the pick attack response.

[kerryandray]

Hey, Transmogrifox.

Thanks so much for the response (and sorry for the late reply).  I thought the extra testing you were talking about had to do with determining how to tweak the Mutron circuit just right, but that was in relation to another project of yours. 

And, yes all that you said was very helpful.  I have read the Neutron instructions and the bit about Rx, and also read the Mark Hammer article on Envelope Filters, but admit some of it goes over my head.  So thanks for clarifying.  Also, thanks for the very clear explanations of what affects sensitivity/range versus depth.  I'll have to play around a bit with different resistor values and see the effects of the changes for myself.

So...I will slap in the vactrol I made and see what happens.  And I will refer back to your post if I need to tweak the sensitivity/range, depth or pick attack response.

Once again...thanks so much for your in depth response.  I hope it helps other newbies like myself as it has certainly helped me.  :-)

Ray