Roland Jazz Chorus JC-20

[Yazoo55]

OK, I’ve breadboarded the preamplifier section - it just about did my nut in, but it’s worth it when it finally works. I had real problems with the chorus and I spent literally days getting it working. I couldn’t get the original LFO to play nice with the MN3207 and in desperation I grafted the chorus section from another pedal in. I finally realised the problem was the bias voltage going into pin 3 of the MN3207. In the original circuit there is no way to adjust this. I added a bias adjuster and got it working yesterday - finally. The advantage of my modified circuit is that it includes both a rate and a depth pot allowing deeper control of the chorus.

By the way, I was wrong about those resistor values. The stated values do work. I had mis-wired the very first op amp stage.

Here is a picture of my very hairy layout. Instead of using a Zener diode to limit the chorus chip to 9 volts, I used a voltage regulator (on the middle board). I tried using a Zener but it got very hot.

[brianhujung]

Awesome!
Rate - depth knob will be very nice improvement over the original..
For regulator, I think 78L09 / LM317L will be sufficient for this.
I Look back on the schematic for section that goes to pin 3 of MN3207 and here's my finding :

[Yazoo55]

Thanks, I’ll change that resistor value. I’ve ordered in some TDA2030s which are supposed to be good substitutes for the upc1238s so the power amp is the next stage to try out.

[Yazoo55]

I’ve updating my Kicad schematic with the new chorus. I’ll start laying out the pcb next.

Looking at the photos you provided, I noticed the distortion potentiometer, like the volume potentiometer, is marked “104B” which must mean 100k linear. The schematic gives the distortion pot as 25kA. You said there are two different versions of the board and I can see differences between the photos. Any suggestions on which value to use? We can always alter parts values later and it won’t interfere with the pcb layout.

The only other query is the Zener diode value, D6 in the mute circuit. I just wish I had the knowledge to work this out myself but it’s beyond me.

[brianhujung]

I’ve updating my Kicad schematic with the new chorus. I’ll start laying out the pcb next.

Looking at the photos you provided, I noticed the distortion potentiometer, like the volume potentiometer, is marked “104B” which must mean 100k linear. The schematic gives the distortion pot as 25kA. You said there are two different versions of the board and I can see differences between the photos. Any suggestions on which value to use? We can always alter parts values later and it won’t interfere with the pcb layout.

The only other query is the Zener diode value, D6 in the mute circuit. I just wish I had the knowledge to work this out myself but it’s beyond me.

I would gambling on the B100k on the distortion potentiometer because it's written both on the potentiometer and the board..

Dist-B100k.png (457.51 KiB) Viewed 5550 times

D6 on the schematic is equal to D8 on the Board. After reviewing some pics, I would suggest to put 1S133 instead zeners on this position because it has same physical appereance with D11.

D8-D11.png (460.23 KiB) Viewed 5550 times

here's another clear picture to compare D4 which is Zener (has black band marking) vs D5 schottky 1S133 ( which has yellow bands marking)..

Hope this helps!
Cheers,
Brian

[Yazoo55]

Thanks. Are there any other pictures for the rest of the potentiometers? It would be good to confirm them as well. I’ve looked at the schematics for the JC-120, trying to make a comparison, and they are very different.

[brianhujung]

Today is my lucky day.. Here's the latest picture I found on the web :
D8 is for the led, D9 are the "zener" on the muting section.
It also give some hint about the value of R30 which is probably 680 ohm.
Also R41 probably 68R or 680R

Cheers,
Brian

[Yazoo55]

Excellent, this really helps.

[Yazoo55]

Would you like the Kicad files sent to you?

[Yazoo55]

I have assembled the main board without the power chips for the moment. I tested it without the BBD chips to make sure the voltages were OK and I’m glad I did. I used a 12V bipolar power supply I had and while the other chips are getting about 9.3 volts, the sockets for the MN3207 and MN3102 are measuring 10.3 volts which is too high. I went back to the original design for VDD after seeing a video explaining how this part of the circuit works - it is designed to reduce noise. I used a 9.1 volt Zener. I will experiment to see how to get the voltage down.

The good news is the rest of the circuit works fine. I used an audio probe on pin 3 of the MN3207 and the signal is very strong.

[Yazoo55]

I connected the board up to the full transformer, a 2 x 15V toroidal and the voltage was too high, about +20v/-20V. I think I’ve cooked the TDA2030s in the process. I have changed over to a 2 x 10v transformer I already had and the voltages are lower, still too high for the MN3207.

To sort this out, I added a stripboard with a 78l09 after the capacitor multiplier section and I now have 9.2V for the BBD. I used an audio probe on where pin 1 of the TDA2030 would be (after removing the fried(?) ones and I am getting good signal on both the chorus and normal channel.

The puzzle is how Roland got the correct voltage for the BBD with the existing circuit.

[Yazoo55]

I have got it up and running now. I replaced the TDA2030s and both channels are working with chorus on one channel and the straight signal on the other. I changed the capacitor values C22 and C46 which connect to the 1K5 resistors on pin 2 of the TDA2030s to 22uf (from 4.7uf on the schematic). This is the value used on the data sheets for the TDA2030 and the upc1238.

I am in the process of boxing it in a 2U rack - lots of drilling!

[brianhujung]

Glad you made it works man!

Initially, I got mine works by slapping some regulators + CE-3.. but I never wrap it into final one because the curiosity about how they biasing the BBD is killing me lol.. I can't even make the LFO working.. all the chorus section is total mess.. So there she is sitting in the corner gathering dust waiting for the puzzle to be solved..

Anyway, for the power amp I'm using LM1875 running on +/- 21VDC.. TDA2050 also will be fine for such voltage.. I changed the power section as per datasheet because the original values are unstable on my board ( got some motorboating, probably because of routing issues due my lack of PCB routing skills lol).
It was fun project, but after some evaluation there's too much efforts on this JC-20, and yet it lack some of flexibility that I need.
I probably will start another Jazz Chorus that have more feature, one thing for sure is I'll keep these the power section on JC-20 because it's ridiculously simple and more than enough for my needs..

Keep us posted!

[Yazoo55]

Sorry it didn’t work out for you. I can only guess that they selected specific MN3207s to work in the existing circuit. I don’t understand why they didn’t put in a bias adjuster. I must have spent nearly 2 weeks on that when I breadboarded it before I gave up and grafted in a different chorus design.

I have got mine boxed now and it would all be finished except that I have had to order another illuminated switch because I managed to connect it up wrongly after all that. I also blew 2 fuses. I used 500mA fuses but it turns out the current draw is higher. I ended up with a 1A fuse on the board.

Best wishes!

[Yazoo55]

I’ve finally (almost) finished. Over Christmas I put an 8 ohm speaker I already had into a diy cabinet made out of an old melamine cupboard door. I’ve set this up with another speaker so I can sit in the middle and I am very pleased with the sound. It doesn’t need any extra pedals or even reverb and sounds really good just by itself for that clean chorus sound. It has been well worth all the work and thanks again Brian for all your help.

[Lani]

Was just looking at the schematics on that website and it looks like he missed a pulldown resistor off the MN3207 outputs (7&8). Maybe it's the missing R49?
(Thanks to Dirk who had just spotted a similar mistake on one of my schematics)

Edit .... If I had noticed there was a second page to the thread would have read that you already figured it out ...

[brianhujung]

I’ve finally (almost) finished. Over Christmas I put an 8 ohm speaker I already had into a diy cabinet made out of an old melamine cupboard door. I’ve set this up with another speaker so I can sit in the middle and I am very pleased with the sound. It doesn’t need any extra pedals or even reverb and sounds really good just by itself for that clean chorus sound. It has been well worth all the work and thanks again Brian for all your help.

All the hard work on this project is yours, I did a very minimum efforts on this project, only assist with limited gutshot photos.. Right now I put this project on the drawer until someone can figured out how the chorus works.. I will post if I found new leads or updates..

A little bit sidetrack from JC-20, I collaborated with some local builder here and we came up with another Jazz Chorus.
Preamp is based on JC-55. Chorus is modified to works with 3207-3102. Distortion and Chorus all switchable via front panel or footswitches.
The most important addition is stereo effects loop ( single send-stereo return, all buffered). The power amp itself is in stereo configuration, so it's a waste if it can't accommodate stereo pedals..

We tested it using pairs of TDA2050 and I'm really happy with the result. I don't know how to describe it but there's sort of spatial sensation from this stereo configuration that very ear pleasing, especially with stereo modulation pedal.. And the funny thing is, unlike JC-20 or JC-120, the distortion on JC-55 is quite decent. Maybe because the distortion circuit is pretty much basic like MXR dist+ and also we are using marshall 1922 cabinet..

For the prototype we use a pair of TDA7052A on the power amp because his 3d printer can only print 19cm x 19cm.. however it still produce good result and now we have great bench amp..

[brianhujung]

Was just looking at the schematics on that website and it looks like he missed a pulldown resistor off the MN3207 outputs (7&8). Maybe it's the missing R49?
(Thanks to Dirk who had just spotted a similar mistake on one of my schematics)

Edit .... If I had noticed there was a second page to the thread would have read that you already figured it out ...

Hi Lani, did you notice any other suspicious thing on the chorus section? The signal path, value or anything? I'm really open to any suggestion on how to make the chorus section works..

[Lani]

It still looks like these 2 resistors are missing from your schematic (yellow square R49 and R51 both 68k). If I were to guess they are the pulldown resistors for the 3207, one coming off each output to ground, then they go through the 6k8 resistors.... Putting a bias control on the input couldn't hurt either, as was stated earlier.