has anyone here successfully cloned the Roland AP7 Jet Phaser

[njkmonty]

http://aussiechat2000.com.au/SP/AP.html

[craigmillard]

Hi,

I have cloned this pedal but havent put the documentation together yet as i have a batch of stuff i want to release back to the community! I even have a few PCBs from my testing:)

Give me a few weeks and ill put all of it up! So far i have:

Roland JetPhase (few PCBs)
Maestro MPF-1 (few PCBs)
Prunes and Custard
Ultimate Octave (few PCBs)

Im also building a mini stack amp with high voltages which im hoping to put up here. each section is a 1444-16 enclosure
So far this has:

Alembic/Fender clean preamp
JCM800 preamp
Uberschall preamp
2W 6n6p/ecc99 poweramp

Hopefully get the stuff up on a website soon!

Thanks
Craig

[njkmonty]

ive printed out as much pcb layout schematics parts list etc , but wasnt sure if it was really important having matched obsolete jfets etc.

so you were happy with your clone results?
i just finished a ce 1 and stumbled across these other Roland early effects

[Mark Hammer]

I have a board etched and stuffed, but wiring up and troubleshooting is waiting until after retirement.  The 6-position mode switch will be replaced with something more sophisticated at that time.

[njkmonty]

were you using the old 2sc....  transistors / jfets  etc  or planning on using current equivalents?

[craigmillard]

Here is the schematic and board layouts although i will put them in a better format when i get the chance:




Im using it in the band at the moment, very lush phase:) sounds good! the fuzz is very harsh not really my thing but can see its uses!

I used roughly matched 2n5457's and it sounds good..

[njkmonty]

excellent

[njkmonty]

I have just finished this build using
2n3904 instead of 2sc373
2n5089 instead of 2sc1000gr

used required 2sk30  from tayda,

http://www.taydaelectronics.com/2sk30-2sk30atm-y-n-channel-50v-6-5ma.html

which i matched for the phase stage  (-1.7v)

once completed , i liked the jet distortion component but could hardly hear any  phasing , even after playing with the trim.

After reading  different posts by Mark Hammer and reading up again on Geofx jfet matching web pages, i finally understood that the jfets acts like variable resistors, turning on and off pending the voltage

but i wasnt quite sure what to do.

anyway in the meantime i decided while i was debugging i wanted to see the lfo so i knew what it was doing and poked around and managed to find a spot that an led would respond too.

with my surprise once i did that the phasing could be heard slightly more.

I read some more not knowing really what i was doing ( like building an amp for the first time!) i tried to find a way to play around with the voltages being sent to the phasing jfets.

i basically connected an alligator clip to the trim pot and sent it to a 25k pot and then to ground.
to my surprise bleeding some of the voltage to ground , then made the phasing much better. 
after fine tuning , i used 2 resistors  to give me the same value ( 14.25K)
now when i turn the trim pot it behaves like a manual pot, adjusting the spectrum of the phasing ie turn it up it phases at higher frequencies etc.

im quite happy with the result, however i was wondering if anyone could maybe help explain what ive done, and if maybe i should do something a little differently ?



I also used a 3p4 t rotary switch instead of the 4p6t. I omitted the jet tone " bass and treble " switching and implemented a Mark Hammer suggestion from another post  removing R13 ( 100k) and replacing with a 100K pot   ( creating a tone pot of sorts) making it a bit more versatile.

I also running of a simple Lt1054 surge pump circuit for the bipolar 9v

im trying to link some pics however photbucket is down for me, will try again shortly


i can upload a video if there is any interest

[Mark Hammer]

First, congratulations!

Second, much respect for completing it.  Not exactly your mother's two-transistor fuzz!

Third, I'm pleased that the mods I suggested are working out nicely.

Fourth, I see that there is the same sort of network around each JFET that one sees in the MXR Phase 45 and other similar units.  The network is intended to reduce JFET distortion, as illustrated here: http://www5b.biglobe.ne.jp/~houshu/synth/PhaseFet0205.GIF

As I've noted a few times in past, Mike Irwin* conveyed to me that, while the network did indeed reduce distortion, it did so largely by moving the threshold for distortion onset upwards.  But rather than gradual onset with increasing input-signal amplitude, as in the standard non-network-supported JFET to ground (in a Phase 90), he found there was a sharp transition from clean to dirty when threshold was reached, and indicated to me he preferred the way in which gradual transition sounded.  The caveat is that Mike tends to operate in the world of synths, where typical signal levels will be hotter than guitar.  So his preferences may be specific to synth signal levels.

That said, one starts to see a pattern in the relationship between use of the distortion-reducing network, and the overall design of the phaser.  The Phase 90 does not use the additional network.  But it also uses a unity-gain buffer at the circuit input, and applies a little bit of gain at the output mixing stage.  In contrast, the Phase 45 applies gain at the input stage, and uses passive mixing on the output.  Here, with the AP7, we also see the distortion-reducing network.  We also see passive mixing at the output (R22/R23 in your drawing), PLUS there is a distortion stage prior to hitting the phase-shift stages.  So, in keeping with the P45 structure, gain-at-the-input and passive mixing at the output, is compensated for by use of the network around the JFET to raise the clipping threshold.  There IS a pot to attenuate the clipping-stage output to "safe" levels.  But since any given musician is entirely at liberty to crank the distortion and use the AP7 like a modulated fuzz, if they wish, the JFET network is still needed as a safety precaution.  I suppose it should also be considered that, if the distortion level from the fuzz stage is severe enough, who is going to notice the sudden onset of clipping from the JFETs, right?


*If you have contact info for him, please let me know.  I don't know how or where to reach him these days.

[duck_arse]

ifn you measure your DC voltage at the wiper of RV4 with and without the extra 14k, you could probably rejig the 100k and 22k resistor values to get the trimmer voltage range into the good voltage zone. (it is not unknown for the bias pot, once range limited, to be panel mounted, but that is a different thread.)

that's a nice looking thing, tho. where did you get the board/layout?

[Mark Hammer]

I don't mean to criticize.  And I think the board looks great and completely sympathize with the urge to keep things at a low vertical profile by bending the socketed JFETs over.

But tell me: where are you going to find an enclosure that accepts a board of those dimensions but won't have enough room for the JFETs to stand up straight?! 

[njkmonty]

where did you get the board/layout?

http://aussiechat2000.com.au/SP/AP.html

this is where i got the layout

top left hand side in blue is the links for the project

ifn you measure your DC voltage at the wiper of RV4 with and without the extra 14k, you could probably rejig the 100k and 22k resistor values to get the trimmer voltage range into the good voltage zone. (it is not unknown for the bias pot, once range limited, to be panel mounted, but that is a different thread.)

will give it a go and report back!

I don't mean to criticize.  And I think the board looks great and completely sympathize with the urge to keep things at a low vertical profile by bending the socketed JFETs over.

But tell me: where are you going to find an enclosure that accepts a board of those dimensions but won't have enough room for the JFETs to stand up straight?! 

all good, actually its hard to see in the pic before, but i have cut and soldered into the sockets the non phasing transistors.
the reason i bent the other ones in the phasing section was,  prior to that photo i had all the off board wiring going everywhere and the jfets where getting knocked out when i adjusted the rotary switch or pots!
once im happy with them i will cut down the size then solder a leg into the socket!


i will do a little video after playing around with trim, 
more to come

[njkmonty]

Hey  craigmillard!!

i just noticed on your schematic a final transistor stage that you added on?   just curious why you did that?

[anotherjim]

Interesting projects and good work getting this unusually complex (for a stompbox) thing working. One is on my to-do list.
As for the "Jet distortion", try it with a bass.

I do know that many of the originals hit landfill due to a fault too complex for many people to fix - only due to lack of easily available schematics back then and that they used a version of the LM324 with an unusual part number - LM2902. I guess many assumed it was a special chip.

Anyway, the original battery power scheme has no reverse protection - just touch a battery snap the wrong way around and at least one of the quad amps can blow. I would advise fitting reverse protection.

[Mark Hammer]

I finished up a 6-stage version of the Phase 90 for myself the other night (on perf, so don't ask me for a layout), with a few other mods (variable resonance, variable offset/range, variable sweep width).  And after viewing a Youtube or two of the AP7, decided to patch in a homebrew slightly altered Distortion+ in front of the phaser.  I won't say that it sounded exactly like the AP7, but it sounded pretty close, to my ears.  I imagine some of the audible difference was the use of 8 stages in the AP7, vs 6 in mine, and the nature of the tone control in the "jet" section of the AP7, as well as the tone shaping of the feedback in the AP7.  But the difference was not huge.

[njkmonty]

i was wondering if i could tap an output from the middle wiper of  the jet pot to an external jack, perhaps try the jet section in front of a flanger?   the jet section does sound a little different to other stuff i heard.
i listened to my video, and i must say it sounds much  better in the flesh, i notice too when i change the rotary switch  between the different depth settings it also lowers the sweep range frequency, when its on the xtreme jet setting to my ear its cycle range is in higher frequencies

maybe i could get carried away and build another and make a Bi phase jet phaser!  with Through zero phasing

no  only kidding

[Marcos - Munky]

Did you drew the pcb layout by yourself? I didn't found it at the link you've posted, just found the parts layout. After watching your video, I'm thinking about if I should add it to my projects list

[njkmonty]

ok here is everything i could find,, cant remember where i found them!










ive got some dropbox stuff too coming....

it  has the pcb image etc

https://www.dropbox.com/sh/y9hprqmicvs8a3v/AAAWvRO4T1xvT5iWfIWTbbvha?dl=0

hop that helps