Fender Chroma Polaris

[Jim Alfredson]

I picked up two Chroma Polaris as part of a large lot of synthesizers last year. The Polaris released in 1984 after development delays. Had it released a year or two earlier, I think it would've been a hit. It has this amazing ability to sound both analog and kind of digital at the same time, by which I mean you can get some really nice FM-ish tones out of it but still with that analog warmth.

Anyway, both Chromas are in generally good shape but neither one would auto-tune properly. First order of business is always checking the power supply rails and they were both fine. Also, both Chromas suffered from the common issue of the membrane panel buttons failing. This is due to the material used in the very thin ribbon cables that connect the panels to each other. The plastic in those thin 'cables' becomes extremely brittle and breaks and the traces seem literally just painted on.

Thankfully one of the head designers, Phil DeRocco, is active in the community and makes new membrane panels to the exact specifications, which he sells on eBay. And the ribbons are much better quality. They should last several lifetimes.

So far I have installed the new membranes on one of the Chromas, but I have another set for the second Chroma. That particular Chroma worked fine after I re-seated some chips. The auto-tune worked perfectly and it sounded great. I was playing it every night for a month and it worked flawlessly.

 

Then I put it away for about a month, until I got it out last week again to mess around with. Unfortunately, it started freezing after about 10 minutes. A reboot didn't help. I had to wait a good 30 minutes or more and it would come back to life and then freeze again.

I brought it out to the bench immediately after another freeze and surprisingly it booted up just fine. No need to wait 30 minutes. Left it on for an hour and it worked great. So I took it back into my studio, put it on a stand, and turned it on. 10 minutes later it flipped out again.

I determined that the cheap keyboard stand was allowing the unit to flex a little bit, and that flexing was causing issues. I figured I needed to deep clean some sockets.

 

So I spent some time doing some more deep cleaning and found that the socket for the DAC was especially dirty, along with some others. Some contact cleaner fixed that right up. But a new problem arose. Voice 5 wasn't responding to auto tuning and thus was getting automatically shut off.

 

Manually turning the voice on revealed that OSC 1 of that voice was totally dead. OSC 2 was working but of course out of tune because it relies on OSC 1 also working to tune.

 

Swapping CEM3374 chips between the non-working voice and a working one yielded no change. So thankfully it wasn't a dead CEM chip.

I initially suspected C618 because my limited understanding of how the CEM3374 works is that it needs those two external capacitors to start oscillating. I tested both caps in circuit by removing the CEM3374 from its socket and testing across them, since without the chip in place they both just go to pins 13 and 15 (now empty) and then to ground. So that's just like testing them out of circuit and they both tested fine.

 

Then I went on a wild goose chase to try to find the culprit. In so doing I found another fault; a bad resistor on the CV input to OSC 1. If gently pried, it would lose contact. Bad solder joint? No. I reflowed all the solder joints on the chip socket and surrounding components, because in my testing, OSC 1 would sometimes work, sometimes not, sometimes come back when gently pressing around the area, sometimes not.

 

Bad socket you say? I replaced the CEM3374 socket. That seemed to solve the problem. It worked for a good hour and then died again.
 

Long story short, after days of testing and probing and thinking I had solved it, only for OSC 1 to die again after an hour or so of the unit being on, I eventually pulled C618 and swapped it with it's doppleganger, C718 from voice 6. I turned on the Polaris and the problem was still there... but this time it was on voice 6! So it was that damn cap all along. Even though it tested fine in circuit and even out of circuit, it would go wonky after being on for awhile.

 

 

I replaced it with what I had on hand; a 1000pf mica 600v cap. A little too big but it works. I ran the Polaris for three hours with no glitches or faults of any kind.

That was a few weeks ago.



Two days ago I set it up again and started playing and it locked up. Again. Turning it off and on it came back and voice 5 is still functioning fine. But then it locked up and wouldn't come back up after a reboot. But moving the case made it come back to life. So there's still something intermittent.

The only chips I didn't reseat are the SRAM chips, because to do so means losing all your patches. But I decided to do it anyway. To the best of my knowledge, there is no way to bulk dump the patches via SYSEX, though you can bulk load them. This means I needed to save the patches via audio out of the TAPE connector or the SYNC connector on the back. I set up a cable to record the digital audio signal from the SYNC connector to my computer and turned the Polaris on and realized that half of the patches were gone.

So that just confirmed my suspicions; one or more of the RAM chips have a dirty socket and it's losing connection to the bus when the unit is on for awhile, which causes it to freeze. 

Well, since most of the patches were gone, I decided to not worry about the rest and removed all the SRAM, tested it all in my Retro Chip Tester. Every chip tested fine. I cleaned the sockets and put them back in. While I was in there, I adjusted the analog and digital +5vDC rails as outlined in the service manual. The rails were just a tad high at around 5.02vDC. I adjusted them to exactly 5vDC.

Once you remove the SRAM like that, you have to recalibrate everything, most of which is done via the Chroma's software. First you have to clear all the memory and then start setting up various parameters. It doesn't take long and there's a step by step guide in the service manual. So I did all that but the tuning was way off on two or three voices. Especially in the upper end. Hmmm...

I played with it for an hour, trying to get it to tune properly and it just would not do it. Then I thought about that voltage change. I admit, I don't know how accurate my DMM is... so maybe I set the voltages too low? I put them back to where they were and voila, everything tuned up just fine.

I had it on for two hours last night and no freezing. Hopefully this time it really is fixed!

[Doerfler]

On 6/15/2024 at 4:23 PM, Jim Alfredson said:

Hopefully this time it really is fixed!

Inquiring minds want to know?    

[Jose EB5AGV]

Pretty interesting and detailed troubleshooting, at least for electronics freaks as myself 😅

 

Never give up, that is the golden rule to success! 🏆

[The Real MC]

After a long hunt, I narrowed a Memorymoog autotune problem to the Z80 CPU.  The rest of the synth was working fine, seems that autotune uses a specific register on the Z80 that was defective.

Lesson learned: don't rule ANYTHING out.

Another intermittent embedded system problem turned out to be a PC board support that was shorter than the rest.  It was flexing the board just right.

 

Lesson learned: not EVERY cause of a circuit malfunction is of an electronic nature.

[AUSSIEKEYS]

Hmmm. I did repairs on mine in the late 80s i guess. Id bought it used from the Chroma importers in Sydney but with a caveat it had problems even though theyd serviced it. But i knew i was up for opening it up as they were honest. I wanted one and i was there and it was cheap enough for me.

 

I pulled it apart and reseated the cables to the keybed and various other things and got it to work much better but it always seemed weird. 

 

I didnt have to do the membrane switches as it was not very old but nevertheless a faulty synth that was rejected by the importer. 

 

All i can really say is ive owned one as the playing experience was just not enjoyable to even remember what it sounded like.  

 

I didnt know they could twist.

 

I at least got it working better for the next custodian who i told it was needing further attention. I think i at least broke even. 

[Mike a]

Is there a way to mod the Polaris to receive and play chords via midi from my Casio PX 5s?

[eliot888]

On 8/5/2024 at 6:19 PM, Mike a said:

Is there a way to mod the Polaris to receive and play chords via midi from my Casio PX 5s?

It's highly unlikely and not worth the effort, BUT you could use a DAW to play your Casio PX-5s, process the audio, and then output it to the Polaris. This would give you much more flexibility but requires additional equipment and software.

[Jim Alfredson]

On 6/18/2024 at 3:42 AM, Doerfler said:

Inquiring minds want to know?    

Nope. 
 

Turned it on for the first time since that post the other week. It will not successfully auto tune. Everything is way out. 
 

Very frustrating.