electro-music.com   Dedicated to experimental electro-acoustic
and electronic music
 
    Front Page  |  Articles  |  Radio
 |  Media  |  Forum  |  Wiki  |  Links  |  Store
Forum with support of Syndicator RSS
 FAQFAQ   CalendarCalendar   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   LinksLinks
 RegisterRegister   ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in  Chat RoomChat Room 
 Forum index » DIY Hardware and Software
Yet another Moog 914 clone...
Post new topic   Reply to topic Moderators: jksuperstar, Scott Stites, Uncle Krunkus
Page 1 of 1 [14 Posts]
View unread posts
View new posts in the last week
Mark the topic unread :: View previous topic :: View next topic
Author Message
dingebre



Joined: Aug 10, 2008
Posts: 252
Location: Salt Lake City, UT USA

PostPosted: Tue Jul 02, 2013 12:31 am    Post subject: Yet another Moog 914 clone... Reply with quote  Mark this post and the followings unread

OK, so I built 3 of Yves Usson's EXCELLENT 914 clone modules:
http://yusynth.net/Modular/index_en.html
Thank you Yves.

I purchased 2 sets of PCBs for Jurgen Haible's version:
http://www.jurgenhaible.org/ffb/fixed_filter_bank.html
Thank you Jurgen, rest in peace...

and I've studied the version Carsten Toensmann created:
http://www.analog-monster.de/mmt914_en.html
Thank you Carsten. Also, Carsten is offering real inductors of custom values, particularly the ones needed for the Moog 914.

Despite all these excellent and well done versions, I still wanted one that was laid out to fit in euro rack panel, pots on the PCB to minimize flying wires, and the ability to easily use real inductors should I find a source... I DID find a source! Carsten Toensmann will hand wind inductors! Please check his site for information and all that.

So, now more than ever I want a PCB set that has the Moog I/O circuits and real inductors. There may be a difference because I believe the Moog inductors were toroids and I believe Carsten's are cylindrical. Other more bright people than I can comment on whether or not that matters.

I'm struggling with the idea of offering these for general consumption because I do not want to step on anyone's toes, especially Yves, Jurgen, and Carsten. While virtually all of the design will be unique, it will also resemble Jurgen's and Carsten's closely by necessity (at least at the schematic level not at all at the PCB layout) and borrows ideas from Yves' design. So, if anyone thinks I'm out of line, I won't offer the PCBs, but will post progress and the learning experience at the least. I'm not making money on this stuff, believe me nor do I want or need to (I still have several thousand dollars of Synthasystem PCBs gathering dust in my basement to prove it Smile ) I just want to share some fun stuff and maybe make back some of the hard expenses of getting a PCB made and in hand... Back to the project.

So, understanding not everyone will want to buy the inductors or even like them, I also decided to have some type of active filter cells or simulated inductor based filter cells. I am pretty convinced I want to use an Antoniou inductance simulator [eg: Sedra and Smith, "Micro Electronic Circuits", Oxford University Press, 2010. Starts on page 1285] Jurgen took this approach as opposed to the Deliyannis multi-feedback structure that Yves used.

So, the ultimate question is what were the "selected" capacitors on the original 914 schematic? With some help from Yves (thank you again), some careful math, looking at the details of Jurgen's design and Carsten's design, and Yves' design, I think I have a handle on what the design goal was for the capacitors.

Please, please, please enlighten me if I am off base here. Selecting the capacitors ultimately has the biggest effect on the sound and behavior because all the other values are specified, and I want to get this right.

It appears the two sections of the cascaded bandpass filters are not tuned to the same center frequency. When I tune them to the same center frequency in SPICE, I get a response which doesn't resemble the SPICE simulations of the real inductors, or Yves' active cells, or Carsten's real inductors, or Jurgen's design either with simulated inductors or real inductors, OR the response published by Moog (see http://moogarchives.com/ Drill down to instruments then modules, then 914 and scroll down for the chart). Yes, I've spent way too much time puttering about in SPICE with this thing.

As of now, I am convinced the design goal was to set the center frequency of the first section in the cascade, with the higher inductance, to a frequency about 5 to 6% lower than the desired center frequency for the given cell and the second section in the cascade to about 5 to 6% higher. The result is a bandpass which has a wider pass band, but the desired center frequency. With these capacitor values, the response is very similar to Jurgen's and Carsten's. There are some subtle differences with Yves', but visually, it's close. Unfortunately, I can't hear it til I build it!

Bandpass done which leaves the high pass and low pass cells. Now, the high pass cell and low pass cell are different. The high pass cell is easily done with the above mentioned simulated inductors (or of course real ones). The capacitors can be selected to get a cutoff of about 7000 Hz by Yves' website or about 10000 Hz by Jurgen's or Carsten's designs (if I trust my math). It looks like both cells are tuned to the same cutoff frequency in this case.

The low pass requires a 'floating" inductor, that is, the coil is in series with the signal path and does not have one leg tied to ground. So, the low pass needs to be an active filter (Jurgen and Yves use a similar structure with different values for components) or I've read where using two Antoniou GICs back to back will simulate a floating inductor. So, 4 op amps instead of 2. I like the idea of inductors, real or simulated, so I'm tinkering with a simulated floating inductor. We'll see how it works out... I think the low pass cells shared the same design philosophy as the band pass cells. That is the first section has a lower cutoff, about 72 Hz, and the second cell has a higher cutoff, about 105 Hz for an average of about 88 Hz.

For the input, I'm going to have the Moog original, but also a transistor one with a bipolar supply source selectable with a DIP switch(s) on the PCB.

For the output, I am borrowing the idea of even and odd cell outputs from Yves because it is one of my favorite features of Yves' design; brilliant Yves, brilliant.

These outputs will have a simple non-inverting amplifier section with an active gain control.

The combined output will have the original Moog output amplifier section as well as the same "modern" amplifier section. Which one is in use will be selectable with a DIP switch(s) on the PCB.

Jurgen designed a faithful output amplifier section which I may just use rather than re-invent the wheel. A dilemma.

The input section will also have an active attenuator which will have it's own input jack to attenuate some of today's higher level signals particularly if you are using the Moog amplifier section. The design will be such that this input jack is switched out when you put a plug in the "normal" input jack.

One thing of note. The capacitors used really have to be selected. To that end, I will not specify capacitor values on the PCBs, but will have copious notes on how to calculate the values depending on the value of the simulated or real inductors.

I will do the same for the capacitors in the Antoniou GIC inductors, too. That way, you can make it any way you want it to be.

I will have some suggested values you can start with, in particular for the simulated inductors. But, to get that real Moog sound, you will need to break out your calculator, and with real inductors, an inductance meter to get the capacitors right.

OK, done for now.

I do not require pre-orders, or anything like that. When I do a project, I like to keep extra around and if I run out of PCBs (assuming they do get offered), I simply order more. I don't buy in a huge batch, typically no more than 50 at a time and I use PCB-Pool.com. They do a high quality job, they do it in a pool so I do save some money, and they accept a lot of different formats, including Eagle, so there you go...

I welcome comments and questions. Remember, I'm still developing the schematics, but the input/output, and the basic filter cell is done, I just have to duplicate and do the layout.

There will be 4 PCBs, uses Panasonic pots, similar to the ones Blacet uses but available at Digikey.com Panasonic #s EVU-E2JFK4B54,
EVU-E2JFK4D54, and EVU-E2JFK4D14. Transistors from Mouser, except the 2N2926 (I got a bunch from Hicham Dalbani at http://www.dalbani.co.uk/forums.html Great source for a lot of harder to find parts)

Other parts:
Inductors - Carsten Toensmann www.analog-monster.de . The only source I have ever heard of or found. I have two sets on the way and will let everyone hear who it compares next to Yves' module mostly because I have two of the three I built to compare with and Yves was able to compare and measure his against a real Moot 914, so it's the best test I can give it.

Everything else is standard type stuff.

David

_________________
David M. Ingebretsen, M.S., M.E.
Collision Forensics & Enginering, Inc.

dingebre@3dphysics.net
http://www.xmission.com/~dingebre/Synthasystem.html
Back to top
View user's profile Send private message
LFLab



Joined: Dec 17, 2009
Posts: 179
Location: Rosmalen, Netherlands

PostPosted: Tue Jul 02, 2013 1:42 am    Post subject: Reply with quote  Mark this post and the followings unread

LOL! I also did a layout for a FFB a few weeks ago, used Juergen's discrete input/driver and discrete output. Mine is going to be 10 band.
All inductors (since I had a bunch of ferrite laying around, waiting for a project Very Happy )
Pots are PC-mount, I used Alps RK11K112. No experience with them, but they do have a nice feel.

Tested the input and output stages and these work like a charm (as you'd expect with something with that pedigree).

Good luck with your project!
Back to top
View user's profile Send private message
dingebre



Joined: Aug 10, 2008
Posts: 252
Location: Salt Lake City, UT USA

PostPosted: Tue Jul 02, 2013 3:07 am    Post subject: Reply with quote  Mark this post and the followings unread

LFLab wrote:
LOL! I also did a layout for a FFB a few weeks ago, used Juergen's discrete input/driver and discrete output. Mine is going to be 10 band.
All inductors (since I had a bunch of ferrite laying around, waiting for a project Very Happy )
Pots are PC-mount, I used Alps RK11K112. No experience with them, but they do have a nice feel.

Tested the input and output stages and these work like a charm (as you'd expect with something with that pedigree).

Good luck with your project!


Thanks! How did you decide what capacitor values you will use?

David

_________________
David M. Ingebretsen, M.S., M.E.
Collision Forensics & Enginering, Inc.

dingebre@3dphysics.net
http://www.xmission.com/~dingebre/Synthasystem.html
Back to top
View user's profile Send private message
LFLab



Joined: Dec 17, 2009
Posts: 179
Location: Rosmalen, Netherlands

PostPosted: Tue Jul 02, 2013 3:18 pm    Post subject: Reply with quote  Mark this post and the followings unread

I am going to use the same values Juergen Haible did.
Bought big polyprop caps. Haven't decided on the type of inductors. I have a bunch of pot-core ferrite, but the original used toroidal inductors (not that it will matter much).

I can wind the potcores pretty easily, but the do need a lot of windings (the > 1H Probably need to go on a transformer core).
On the other hand, the toroidal cores are a royal PITA to wind but need far less windings. Decisions, decisions.
Back to top
View user's profile Send private message
dingebre



Joined: Aug 10, 2008
Posts: 252
Location: Salt Lake City, UT USA

PostPosted: Sun Jul 07, 2013 10:32 pm    Post subject: Reply with quote  Mark this post and the followings unread

OK, so I've been agonizing over an inability to get the AC response for the bandpass cells to match that of the Moog filter (See www.moogarchives.com). I was able to find some photos of two different 914 modules and I noticed an additional resistor in parallel with the inductor/capacitor pair in the second second section of each cascaded filter cell. The value was different in each of the two modules, and I wasn't convinced of the color bands, but I found that adding a 5.6k to 6.8k resistor at this point brought the AC response right in line with the Moog and with Yves Usson's design (which Yves tested against a real 914).

So, the Moog schematic is incomplete and this resistor needs to be added to properly tune the 914 response.

At this point, I'm also going to use two Antoniou GICs back to back to simulate the floating inductor for the Low Pass, so if built without inductors, it will be a fully faithful simulation of the inductors for each cell.

I'm just waiting to get my inductors from Carsten so I can make sure the footprint I'm using is correct and I'll start laying out the PCBs.

_________________
David M. Ingebretsen, M.S., M.E.
Collision Forensics & Enginering, Inc.

dingebre@3dphysics.net
http://www.xmission.com/~dingebre/Synthasystem.html
Back to top
View user's profile Send private message
jmejia



Joined: Mar 12, 2009
Posts: 114
Location: portland

PostPosted: Wed Jul 24, 2013 11:56 am    Post subject: Reply with quote  Mark this post and the followings unread

Lots of great info in this thread - thanks!

Anyone have any insight in to the commercial clones floating around? I wonder if any of them use inductors (seems doubtful), or if anyone knows how they stack up to the other clone designs?

The commercial clones I know of:
dotcom q127
COTK c914b
MOSLAB 914

Maybe there are more? (probably)
Back to top
View user's profile Send private message
dingebre



Joined: Aug 10, 2008
Posts: 252
Location: Salt Lake City, UT USA

PostPosted: Wed Jul 24, 2013 1:05 pm    Post subject: Reply with quote  Mark this post and the followings unread

First, I got my inductors and they are amazingly small and compact. Great work Carsten! This will make it much easier to fit it all on PCBs which can fit in a moderately deep rack. On to PCB layout!

Next, to jmejia, in my research, I don't think any of the commercial versions use the passive RLC topology. But, it wouldn't hurt to double check with each of them. If you want the RLC topology, you will need to either wind your own inductors, or buy them from Carsten, which I highly recommend. Carsten is also selling his PCBs, but I think you will need to add that resistor discussed above to the band pass cells to get the proper response.

Last option, my design has now become so different than Jurgen's or Yves' that I will probably be offering my PCBs if you want a euro compatible design.

David

_________________
David M. Ingebretsen, M.S., M.E.
Collision Forensics & Enginering, Inc.

dingebre@3dphysics.net
http://www.xmission.com/~dingebre/Synthasystem.html
Back to top
View user's profile Send private message
jmejia



Joined: Mar 12, 2009
Posts: 114
Location: portland

PostPosted: Wed Jul 24, 2013 1:31 pm    Post subject: Reply with quote  Mark this post and the followings unread

Great- I'd love to hear a demo - ideally of active vs passive versions Smile

I didn't realize Carsten was selling pcbs.. that's very promising indeed. I wonder what the differences are between his and Yves' version when built up with inducters.
Back to top
View user's profile Send private message
dingebre



Joined: Aug 10, 2008
Posts: 252
Location: Salt Lake City, UT USA

PostPosted: Fri Aug 02, 2013 2:35 pm    Post subject: Progress Reply with quote  Mark this post and the followings unread

I think I have everything mostly pinned down. There will be four PCBs, three will have the filter cells, and one will handle power and I/O. I am using the Moog input and output amplifiers with the choice of using an extra output amplifier with gain control so the levels can be boosted to those in other modern systems, and the choice of using a buffered input with attenuation or directly into the Moog input amp.

It's pretty easy to saturate the Moog output amp, so I added a clipping indicator so that I can adjust the input attenuation to prevent clipping. I also added outputs for the even and odd cells separately, each using a Moog output amplifier to maintain the same signal path as for the main output. The only difference is I added a high impedance voltage follower in front of the Moog amplifier so these extra output stages don't load the filter cells and change the behavior.

I'm attaching the current MultiSim model as well as a shot of the AC response, as well as the Moog published response for comparison. I think it's a good match. The "Q" of each cell is around 2.7 to 3.0 give or take and I think the relative gain and output is good.

I'm just going to be laying out the PCBs now, so prototype is about a month or two away.


simulation.jpg
 Description:
Here is a screen grab of the AC response of the Multisim model for comparison to the Moog published response.
 Filesize:  189.61 KB
 Viewed:  61 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

simulation.jpg



914d.gif
 Description:
Moog published response of the 914 will all pots open
 Filesize:  54.72 KB
 Viewed:  48 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

914d.gif



Moog FFB simulation.zip
 Description:
Multisim simulation. You can connect the GIC "floating" inductor simulation for the low pass filter, or use the SPICE inductor model version. This simulation is to be used "as-is" with no warranty or guarantee it is correct or useful.

Download
 Filename:  Moog FFB simulation.zip
 Filesize:  1.2 MB
 Downloaded:  41 Time(s)


_________________
David M. Ingebretsen, M.S., M.E.
Collision Forensics & Enginering, Inc.

dingebre@3dphysics.net
http://www.xmission.com/~dingebre/Synthasystem.html
Back to top
View user's profile Send private message
dingebre



Joined: Aug 10, 2008
Posts: 252
Location: Salt Lake City, UT USA

PostPosted: Mon Sep 30, 2013 5:15 pm    Post subject: Reply with quote  Mark this post and the followings unread

OK, so an update for anyone who cares.

I've got the PCBs for Carsten's inductors laid out and ordered. I re-did the whole layout in DipTrace. Much better than Eagle in my opinion.

Next, will be a layout for GICs instead of real inductors for anyone who would rather go that route.

So, when they come and I can make time to build them, I'll post the results.

David

_________________
David M. Ingebretsen, M.S., M.E.
Collision Forensics & Enginering, Inc.

dingebre@3dphysics.net
http://www.xmission.com/~dingebre/Synthasystem.html
Back to top
View user's profile Send private message
dingebre



Joined: Aug 10, 2008
Posts: 252
Location: Salt Lake City, UT USA

PostPosted: Sun Mar 30, 2014 9:42 pm    Post subject: Reply with quote  Mark this post and the followings unread

Time for a quick update.

I got a prototype all built and discovered inductors will do what inductors do and pick up every stray EM field in the universe! Smile Seriously, I puttheh prototype in my cabinet and discovered I can't move it far enough away from my open frame PowerOne power supply transformer to eliminate the hum. Sigh...

Despite the excellent inductors from Carsten Tonesmann (many, many thanks Carsten) I find I need to abandon my hope of an inductor clone of the 914. The cost and effort of a new power supply system and/or cabinet and rack redesign is far more than simply using a GIC paradigm which should be a very close emulation of the inductors minus the annoying "cons" of a real inductor.

I have now completed a layout for a GIC based unit using a GIC for each of the grounded inductors and a "back to back" GIC for the floating inductors in the low pass circuit. Circuit boards should arrive next week and I'll post the results. The simulations are very encouraging.

David

_________________
David M. Ingebretsen, M.S., M.E.
Collision Forensics & Enginering, Inc.

dingebre@3dphysics.net
http://www.xmission.com/~dingebre/Synthasystem.html
Back to top
View user's profile Send private message
elmegil



Joined: Mar 20, 2012
Posts: 1345
Location: Chicago
Audio files: 14

PostPosted: Mon Mar 31, 2014 10:19 pm    Post subject: Reply with quote  Mark this post and the followings unread

What's a GIC ?
Back to top
View user's profile Send private message
dingebre



Joined: Aug 10, 2008
Posts: 252
Location: Salt Lake City, UT USA

PostPosted: Tue Apr 01, 2014 8:37 am    Post subject: Reply with quote  Mark this post and the followings unread

elmegil wrote:
What's a GIC ?

General Impedance Converter. There are several circuits to simulate inductors, but a couple of more popular forms are Riordan's simulated inductor and Antoniou's simulated inductor [see Analog Electronic Filters: Theory, Design, and Synthesis by Hercules Dimopoulos, Springer-Verlag 2012]. Basically using two op-amps, some resistors and a capacitor, you can simulate the impedance of an inductor and avoid the parasitic effects of a real inductor. You get the behavior of an inductor without the problems.

Both of these forms simulate an inductor with one side connected to ground. If you put two of these back to back (ground terminals connected), you can simulate a floating inductor.

Jurgen Haible used these in his 914 clone.

See attached image for the schematic of one of my filter cells.

David


GIC.pdf
 Description:

Download
 Filename:  GIC.pdf
 Filesize:  21.81 KB
 Downloaded:  17 Time(s)


_________________
David M. Ingebretsen, M.S., M.E.
Collision Forensics & Enginering, Inc.

dingebre@3dphysics.net
http://www.xmission.com/~dingebre/Synthasystem.html
Back to top
View user's profile Send private message
elmegil



Joined: Mar 20, 2012
Posts: 1345
Location: Chicago
Audio files: 14

PostPosted: Tue Apr 01, 2014 4:28 pm    Post subject: Reply with quote  Mark this post and the followings unread

Excellent, thank you for the explanation. Smile
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic Moderators: jksuperstar, Scott Stites, Uncle Krunkus
Page 1 of 1 [14 Posts]
View unread posts
View new posts in the last week
Mark the topic unread :: View previous topic :: View next topic
 Forum index » DIY Hardware and Software
Jump to:  

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum
emSynth

Please support our site. If you click through and buy from
our affiliate partners, we earn a small commission.


Forum with support of Syndicator RSS
Powered by phpBB © 2001, 2005 phpBB Group
Copyright © 2003 through 2009 by electro-music.com - Conditions Of Use