Roland TR-707 Germanium Diode Replacement

Pulling out your hair? Don't know what to do or where to go? Ask in here.
Forum rules
READ: VSE Board-Wide Rules and Guidelines

If your Help request has been solved, please edit your first post in order to select the Image Topic Icon to let others know your topic has been solved.

Roland TR-707 Germanium Diode Replacement

Postby upunkt » Mon Mar 11, 2019 9:35 am

I'm currently refurbishing a Roland TR-707 Drum Computer. One of the issues is a broken diode, D17. In the service manual it's described as "Ge Di" which probably translates to Germanium Diode. Unfortunately, no other information about the diode is available.

In the SM it's written that one could replace D17 with a jumper and swop R73 (2.2 MΩ) to 4.7 M. I'm not sure if this replacement can be done no matter what or if certain other criteria have to be met. Still, I guess it's just an adjustment of the envelope of the Ride Cymbal. EDIT I think this alternative applies to TR-727, with Chimes output on this channel instead of the Ride Cymbal. Anyway, I just did this mod for the moment and it seems to work without much negative influence on sound.

Is there anyone knowing more about the D17 diode type and specification? I guess I could use a standard AA113 but I'd like to know it more precise. Any help appreciated.

Thanks in advance.
Last edited by upunkt on Sun Mar 24, 2019 2:24 pm, edited 1 time in total.
upunkt
Newbie
Newbie
 
Posts: 11
Joined: Sat Oct 13, 2018 1:51 pm

Re: Roland TR-707 Germanium Diode Replacement

Postby Mooger5 » Mon Mar 11, 2019 7:16 pm

Interesting case. Agree with the alternative values being specific for the 727. Can't think of a reason for the use of a germanium diode other than it could add noise to modulate the VCA. The ride cymbal would sound more realistic. Because of the low forward voltage drop doesn't make much sense as the alternative is a jumper. Likewise the use of a sil diode there just for rectification doesn't seem probable when again the alternative is a jumper.
I'd use an obsolete 1N34a. 0.3V drop, noisy "vintage" and still easy to find.
Herrare umanum est.
Mooger5
Expert Member
Expert Member
 
Posts: 1409
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon

Re: Roland TR-707 Germanium Diode Replacement

Postby upunkt » Wed Mar 13, 2019 10:35 pm

VCA modulation! This sounds very reasonable, thanks.

I checked specs for 1N34a and could not find the 0.3 V drop. What I found is that depending on the manufacturer the diode characteristics vary in a certain range.

What I can get very easy is an AA143 (from DSI) which claims to have a VF of about 0.3 V, depending on current. The 707 is otherwise fully working again so this would now be the last step.

Just for completeness: Anyone being able to contribute information about the original diodes used in a 707 feel free to leave a hint here.
upunkt
Newbie
Newbie
 
Posts: 11
Joined: Sat Oct 13, 2018 1:51 pm

Re: Roland TR-707 Germanium Diode Replacement

Postby Mooger5 » Fri Mar 15, 2019 12:24 am

I think the 1N34as I matched for a ring modulator were fakes. They were probably Schottky diodes.
Now, being pragmatic, if there is no further information on the required component, I´d say differences between germanium diodes are not critical and any one should fit.
I forgot about the exponential response of the diode. It´s what gives lag processors their characteristic curve. A logarithmic decay envelope as opposed to the others that are linear because of the jumper makes more sense than a noise modulated VCA. Especially for the ride cymbal, it will make it sound more natural. The crash being linear just needs to perceively last longer.
Herrare umanum est.
Mooger5
Expert Member
Expert Member
 
Posts: 1409
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon

Re: Roland TR-707 Germanium Diode Replacement

Postby upunkt » Sun Mar 24, 2019 2:09 pm

I tried both AA113 and AA143. Both Ge diodes let the ride cymbal decay smoothly, while the ride volume level was also affected and dropped. Sound was not affected.

AA113: decay a bit too fast but very soft in the end, volume slightly lower than the crash volume

AA143: smooth decay as long as the sample lives, volume slightly louder than the crash

I decided to keep the AA143 and started wondering why Roland did not make use of this decay technique for the toms.

If anyone has spare fader caps, let me know. Otherwise my TR-707 is fully functional again.
upunkt
Newbie
Newbie
 
Posts: 11
Joined: Sat Oct 13, 2018 1:51 pm

Re: Roland TR-707 Germanium Diode Replacement

Postby Mooger5 » Sun Mar 24, 2019 3:45 pm

Glad it works.

upunkt wrote:I decided to keep the AA143 and started wondering why Roland did not make use of this decay technique for the toms.


I think it´s because in 1985 sample memory was expensive. The crash and ride (quijada and chimes) have their own sample ROMs because of their longer release times. The ride is the only that, according to traditional styles, had to hit every 8th note or so, sometimes 16ths, and sound good, not cluttered hence the special curve; while the other three were meant to hit once every two bars or something, just now and then.
The rest of the sounds, toms included, besides having the decay sort of embedded in the samples, all have identical cap and resistor values for the (linear) envelope generators.

Cheers.
Herrare umanum est.
Mooger5
Expert Member
Expert Member
 
Posts: 1409
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon


Return to HELP!

Who is online

Users browsing this forum: Bing [Bot] and 10 guests