TK. Posted July 9, 2007 Report Share Posted July 9, 2007 I used some spare parts and haven't taken care for certain trimpot values or capacitor types. The circuit matches with the schematic in the datasheet. No tricks, no black magic.For best circurity please ask an analog expert, and not me - I'm a novice in this area and the wrong person who could give you the right tips. I'm happy enough that I haven't fried the CEM3378 during the experiments, and I don't want to take over the responsibility that you destroy your own CEM3378sBest Regards, Thorsten. Quote Link to comment Share on other sites More sharing options...
Rio Posted September 26, 2007 Author Report Share Posted September 26, 2007 1. I want to use a TL072 for 2x CEM3378.Which Voltage should be connected to -V and +V pins of TL072 on CEM3378-board? Should i use +12V and -5V or should i use +12V and -12V?2. if i chain the AOUT_LC, then i only connect a flatcable from J2 of AOUT_LC1 to J1 of AOUT_LC2, right? Quote Link to comment Share on other sites More sharing options...
TK. Posted September 26, 2007 Report Share Posted September 26, 2007 A balanced voltage is required for the OP amp TL072, therefore +/- 12V is recommented.2: yesBest Regards, Thorsten. Quote Link to comment Share on other sites More sharing options...
Rio Posted October 16, 2007 Author Report Share Posted October 16, 2007 ok, i'm able to control the AOUT_LC Voltage from SID Application Res and Cutoff endcoders, but how can i adjust the trimmer correct for AOUT_LC module? What should be adjusted first - the gain or the offset - and which values should be set? (I'm not really sure, i think they all work between 2,8V...8,xV at time - they are all standing in middle position).i've connected the CEM Filter Board to the AOUT_LCs, like in schematic of CEM3378... ... but at time i can't hear any sound. It seems that the power voltage is correct on every IC socket. What could be the reason, that i don't hear anything?The "Gain"- and the "PAN"-Input are not connected and opened.TK wrote: I added two trimpots against ground for an additional gain fineadjust. This was especially important for calibration of the four CEM3378 to ensure, that the same software driven cutoff/resonance value will result into the same effect at the analog side....is that really be necessary or would be nothing heard, if i let them opened? Which type of trimpots you use (10kOhm ok)?Greets Rio. Quote Link to comment Share on other sites More sharing options...
TK. Posted October 16, 2007 Report Share Posted October 16, 2007 It's really difficult to help you when you don't exactly describe, how you've connected the CEM3378.You could draw a schematic to demonstrate, what you did. Maybe this give somebody the right hint what could be wrong. And in addition, it would be a useful documentation for other people who are interested in connecting such a filter chip.It makes sense to start with the highest gain, and the middle offset, and thereafter to reduce these settings until you can sweep over the whole range. Yes, for the fineadjust I used 10k trimpotsWhat do you mean with "if I let them opened"? Did you already connect the additional trimpots, and how? Maybe a wrongly connected trimpot is the reason why it doesn't work?Best Regards, Thorsten. Quote Link to comment Share on other sites More sharing options...
Rio Posted October 17, 2007 Author Report Share Posted October 17, 2007 i have done a PCB-Layout for 2x CEM Filter... here is a screenshot and the PCB for eagle light (in zip) in appendix. Please take a look what i've maybe forgotten - i've strictly orientated by schematic.. For all other, who read this: I can't give guarantee for any errors.Power connector from left to right: -12V, -5V, GND, 12VSignal Out from left to right: Left, GND, GND, RIGHTYes i havn't connected trimpots or any signal to "gain" or "pan" and nothing is connected to them. (could that be the reason, that i can't listen to any sound?)CEM3378_handmade.zipCEM3378_handmade.zip Quote Link to comment Share on other sites More sharing options...
Rio Posted October 19, 2007 Author Report Share Posted October 19, 2007 ok.. only that i get it sure and to clear it up:1.What should be connected to GAIN1 and GAIN2 and PAN1 and PAN2.. ? At time nothing is connected to GAIN or PAN2.where should the 10k trimpot added on that CEM3378-PCB?til now, the board (shown above) doesn't work yet. I can't listen to any sound on signal audio out.please understand me...I don't want to experiment, if i'm not sure what i'm doing... Quote Link to comment Share on other sites More sharing options...
TK. Posted October 19, 2007 Report Share Posted October 19, 2007 This really helped to understand what is going on - you routed the audio signal through the mixer and VCA stage - in this case it's obvious that the CV inputs for these components have to be supplied. The datasheet says that the voltage range is between 0V and 5V for these inputs, so just take 5V for maximum volume and full-channel BIn addition the TL072 based output buffer circuit looks strange, I would expect a resistor between the CEM3378 output and the positive input here. You can take the OPamp stage of the MBHP_OPL3 board as inspiration. I would propose to do the first checks by connecting your amp/mixer or soundcard directly to the VCA or VCF output of CEM337810k trimpots: on all CV inputs which are controlled via AOUT moduleBest Regards, Thorsten. Quote Link to comment Share on other sites More sharing options...
Rio Posted October 21, 2007 Author Report Share Posted October 21, 2007 The datasheet says that the voltage range is between 0V and 5V for these inputs, so just take 5V for maximum volume and full-channel Bok you mean i should directly connect 5V to GAIN1 and PAN1 (GAIN2 and PAN2 too), right?In addition the TL072 based output buffer circuit looks strange, I would expect a resistor between the CEM3378 output and the positive input here.that's something i don't understand, because i've build up by using this schematics:http://synthtech.com/cem/c3378pdf.pdfhttp://www.st-andrews.ac.uk/~jcgl/Scots_Guide/datasheets/Opamps/072dia.gifthere isn't any resistor between TL072 negative input and CEM3378 output..on the other hand, you said:between the CEM3378 output and the positive input here....but it's connected to negative input of TL072.so do i need then really a resistor, and if yes, which resistor type/value i've to take?PS:Beside i'll take a look at OPL3 Opamp circuit...Best Greets,Rio. Quote Link to comment Share on other sites More sharing options...
TK. Posted October 21, 2007 Report Share Posted October 21, 2007 Ok, I mixed this with another circuit - let it like it is.Best Regards, Thorsten. Quote Link to comment Share on other sites More sharing options...
Rio Posted October 21, 2007 Author Report Share Posted October 21, 2007 ok i'll only connect 5V at Gain and PAN and pray that them working... Quote Link to comment Share on other sites More sharing options...
Rio Posted October 22, 2007 Author Report Share Posted October 22, 2007 ok, after adding 10k trimpots and doing few fine adjusts... (oh that was a bit complicated) i got 0...5V for every AOUT Channel. The Datasheet of CEM3378 was a bit confusing, because i've readen that the frequency should be controlled between -150mV .. 100mV, but the description tells me too, that all inputs can controlled between 0..5V (*confusing*). So i checked both and 0..5V is working for frequency.I've connecting 5V to GAIN, but there was no need to connect 5V to PAN (should be mixed to Channel A = 0V)ok, only 2 little questions:1. The filter is working fine. I only mentioned that the resonance turns into distortion if i move the resonance knob up to value = 100. Is this normal? (Maybe there is to much gain - i could add a trimpot for Gain too..)2. If everything is silence, i can listen to some background noise (sweeping noise).Is this normal too?Greets and thx's for much support, Rio. Quote Link to comment Share on other sites More sharing options...
TK. Posted October 22, 2007 Report Share Posted October 22, 2007 I guess that you mean self-oscillation instead of distortion. Yes, a trimpot should be added to resonance as well. Set resonance to maximum (255) in the MBSID CS, and tweak the trimpot until the filter starts to oscillate. Thats the best setting.Some additional background noise is normal, but I don't know what you mean exactly with "sweeping noise". If for example a LFO modulates the CutOff frequency, you would notice some "sweeping noise".Best Regards, Thorsten. Quote Link to comment Share on other sites More sharing options...
Rio Posted October 23, 2007 Author Report Share Posted October 23, 2007 Some additional background noise is normal, but I don't know what you mean exactly with "sweeping noise".if i turn up the volume without routing signals through both filter and without activating a LFO, i can listen to 2 types of sound:- continuous repeating noise (sounds a bit like a railroad ;) )- and another small hightone sound, which comes and goes sometimes... and sometimes the small notes rises or fall in a ocillation.....all in very low level, but i'm not sure, if that is normal. The filter is working correct.I've notice, that IMHO the filter will only take effect at CutOff Freq > 1.0V.so i've adjusted C.V. between: 1.0...5.0V for frequencyI noticed too, that CEM filter not really match with parameter of SID-Filter. So SID Filter will not take much effect in lowpass-area anymore, because the CEM filter is too low/deep... seems SID filter will be a bit useless vs. CEM, or?Greets Rio. Quote Link to comment Share on other sites More sharing options...
stryd_one Posted October 23, 2007 Report Share Posted October 23, 2007 self-oscillation instead of distortion. ....tweak the trimpot until the filter starts to oscillate. Oh, if it's self-oscillating well... personally I like my synths to get a bit crunchy if I boost the res enough... I can always turn it down ;)I'm yet to try it on the MBSID but I have patches where the keyscale modulates the filter cutoff and they play the tune with the filter's self oscillation while the oscillators themselves are controlled by other input... Two of them had the resonance distorting a little, one of those had further processing following that, making use of the high frequencies that the distortion introduced.Point is, you might be able to use a little distortion on the resonance if you're creative/weird ;) Quote Link to comment Share on other sites More sharing options...
Rio Posted October 24, 2007 Author Report Share Posted October 24, 2007 so i hope my background noises are ok.Another thing: if i've connected the bipolar psu and if i switch power of C64 optimized PSU on, then i can measure 11V on CV1 (FRQ1 or FRQ2), while core is booting.... after 2 seconds it turns back to 6,3V without activating F2V. After activating F2V in application, everything is in the correct Voltage range. But, would it be alright that 11V are feeding for a short time (2seconds) to the C.V.s of CEM3378, or can this higher voltage level damage the C.V. inputs of filter, while doing this boot..? until now - everything works well.Best Regards,Rio Quote Link to comment Share on other sites More sharing options...
Rio Posted January 23, 2008 Author Report Share Posted January 23, 2008 ok for all whose want to build a board like shown above and want to use with 2x SID, i've checked up that board again. A way to reduce the background noise is: - don't connect ground at all C.V. Inputs (you can let out all that ground wires on board for that inputs), because the ground of your external PSU (+/-12V,+/-5V) is already connected e.g.to AOUT_LC with GND (you need only ground for both CEMs, the TL and SIGNAL OUT on that board).- don't connect GND from SID OUT to GND of AUDIO_INs (to prevent humming noise)Greets Rio. Quote Link to comment Share on other sites More sharing options...
Rio Posted June 23, 2008 Author Report Share Posted June 23, 2008 Now i've solved that problem of swinging background noise:I'd measured a frequency of 700.000KHz on a oscilograph, which was produced by self ocillating of the TL072 and which modulates the whole audiosignal dramatically. For example: if i put my fingers in the near of that TL072... the whole audio signal was swinging...So if you want to build this schematic above, please don't connect all that ground wire to external c.v. inputs and audio signal inputs (add only one single ground connection between both PSUs directly and a direct connection to that board) and add a pair of capacities (2x220pF) parallel to the 22k at TL072. That schematic will give a negative-feedback condition to that disturbing high frequencies.Schematic: 22k +-----/\/\/---+ | | | 220pf | +------||-----+ | | | |\ | --+----|-\ | | \_____| | / -------|+/ |/ That should solve the self oscilation of TL072 and provides a clean filtered sound.(for people, which are interessted in that, i can add a new PCB screen here too)Best Regards,Rio Quote Link to comment Share on other sites More sharing options...
DenDer Posted June 23, 2008 Report Share Posted June 23, 2008 Put it up man ;) Quote Link to comment Share on other sites More sharing options...
ilmenator Posted June 23, 2008 Report Share Posted June 23, 2008 (for people, which are interessted in that, i can add a new PCB screen here too)Yes, I'd be verrry interested in a schematic (even more than a PCB), too! I still have these chips waiting to be used, and I am sure this is the right stuff to keep me from low-light depression once winter approaches here... ;DBest regards, ilmenator Quote Link to comment Share on other sites More sharing options...
Rio Posted June 23, 2008 Author Report Share Posted June 23, 2008 The schematic is the same like in documentation of CEM3378. I only added the additional caps. The PCB shows, where a ground wire should be or not. Besides, i got trouble creating/converting a schematic in eagle....I'll make the newer PCB at evening, uploading tomorrow.Best Regards,Rio. Quote Link to comment Share on other sites More sharing options...
ilmenator Posted June 23, 2008 Report Share Posted June 23, 2008 Thanks Rio,I really do appreciate your effort!Best regards, ilmenator Quote Link to comment Share on other sites More sharing options...
ganchan Posted June 24, 2008 Report Share Posted June 24, 2008 wow, this is the project i need. i'm thinking from too much time to build a group of filters for routing all my synths before the mixer, so i've independent cut/res on each synth!.waiting for news..! Quote Link to comment Share on other sites More sharing options...
Rio Posted June 26, 2008 Author Report Share Posted June 26, 2008 Sorry for delay, .. i havn't time yet, but i havn't forget it. Uploading soon. Quote Link to comment Share on other sites More sharing options...
Rio Posted July 1, 2008 Author Report Share Posted July 1, 2008 ok, i've drawn the newer PCB with the additional caps. I hope that i havn't forget anything or that there isn't an error. It's only handdrawn.Note:The signal out 1x4 pin header: OOOO (Left Ch, GND, GND, Right Ch)Changes:- 2 x addtional 220nF between 12v and GND, and -12V and GND.- 2 x necessary 220pF parallel to the 22K resistors.- additional 10k trimpot to set/tune 0..5V for gain of both CEM- PAN is opened, because AUDIO IN will be connected to both Audio INs- GND_EXT for connecting GND of C64 PSUBest regards,Rio.CEM3378_handmade2.zip Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.