Emulating Roland Chorus
How to emulate Roland Choruses with the Axe-Fx
(from a post from javajunkie from here: http://www.fractalaudio.com/forum/viewtopic.php?f=11&t=10082&start=0)
I have owned 2 Roland Dimension D rack mount units, and for a long time it was my staple chorus sound. I bought one new when they first came out (wow- that makes me feel old!). I sold them some time ago to raise some cash, but after getting my Ultra I borrowed one back for a while. I tested it extensively both by measurement and listening to see if it was possible to nail that elusive ‘D’ tone with the Axe-Fx.
So- I have some information that you guys might find useful……..here goes……
Firstly due to component tolerances and internal trimpots the units can all be slightly different.
The modulation waveform in the one I had (which sounds great) is almost perfectly triangular with only the tiniest amount of ‘dwell’ caused by the very slight 'rounding off' of the triangle points. I have seen a service diagram in the past and the waveform was also shown as a triangle on that. One side of the modulation is 180 degrees out of phase. The speed and depth of the mod. is controlled by the buttons 1 to 4. The unit has 2 mod speeds, roughly a 2 second and a 4 second cycle with 2 different depths. The triangle wave in the Axe-fx sounds very close when all the other variables are matched.
Just as important is the eq applied to the delay. There is a bass roll off before the compander noise reduction. This can be simulated by placing a parametric eq block after the delay, and using the high pass filter at around 120 hz. There is also a ‘brickwall’ filter at around 8khz applied after the analogue delay lines to filter out the clocking frequency. This is not so easy to simulate perfectly as the Axe-Fx low pass filter is not steep enough. However, using the low pass filter section in the parametric gave very similar and good sounding results.
The compander circuit adds a little bit of program dependant distortion, which gives a small amount of extra ‘warmth’ to the delay. If you wanted to go the whole hog you could follow the parametric eq with a Drive block set to a low value of ‘tape’ distortion to simulate this.
The direct signal has a small (1db) bass lift to compensate slightly for the roll off in the delay. I set the chorus mix to fully wet and used a filter block (adding the bass lift) around the chorus and eq blocks for the direct sound.
The ‘D’ handles stereo signals in a special way. With buttons 1, 2 or 3 depressed the delay component for the left channel appears on the right side and vice versa. Inputting a guitar with stereo bouncing repeats sounds fabulous. Signals retain their stereo position but sound lush, wide and spacious. Button 4 returns the delays to their usual L/R position, and combining 4 with 1, 2 or 3 gives more modulation options.
So- you need a chorus block with independently pannable voices. I used 2 voices of the Ultras quad chorus but you could use a stereo delay and add the modulation. Of course, if you only use a mono input a standard chorus block will do. The delay time is not quite the same in the two channels. I think that for all except button 1 one side is about 5ms and the other about 5.5ms in my unit -about a 10% offset. Button 1 selects the slower speed with a longer delay time, but I did not measure this.
Regrettably, for you guys- I did not take notes on the final individual settings needed to exactly replicate the Dimension D sound, but it is possible to get very very close. I really just used this as an exercise to see whether it could actually be done, and to glean information to use in creating, and fine tuning my own definitive chorus. The Dimension D is fantastic, but not perfect for everything as it adds its own colour. Doing all this also confirmed my opinion that the Ultra is possibly the best and most versatile processor available at the moment.
There it is. I hope this helps you Roland chorus nuts out there to nail the ‘D’ sound.
Best Wishes Everyone….