output transistors substitution Roland Dimension D

hello,
I'm building a clone of the Roland Dimension D (SDD-320) and so far the only problem I've found is how to replace the output transistors (Q106,107,108,109).
They used 2SC495 and its complementary 2SA505.
I'd like to use in-production transistors, I don't want to source those old parts.
Is there an equivalent that would reduce to min. the modifications on the schematic?

service manual: http://www.synfo.nl/servicemanuals/Roland/SDD-320_SERVICE_NOTES.pdf
datasheet: https://www.datasheet4u.com/datasheet-pdf/Toshiba/2SC495/pdf.php?id=706036

You can use, for example, BD139/140 or MJE171/181 in that positions without any modifications.

Dimitree said:

hello,
I'm building a clone of the Roland Dimension D (SDD-320) and so far the only problem I've found is how to replace the output transistors (Q106,107,108,109).
They used 2SC495 and its complementary 2SA505.
I'd like to use in-production transistors, I don't want to source those old parts.
Is there an equivalent that would reduce to min. the modifications on the schematic?

Click to expand...

You could use almost anything here, like TIP31/41 or even BC550/BC560 (watch the pin-out though).

In a modern studio, you can just skip the booster transistors. Any opamp will drive 10K loads loafing.

If you truly must drive 600 Ohm loads, use a 5532 opamp. It easily drives 600r direct and does not strain much driving 600r bridge (300r each opamp).

thanks for the replies!

so you suggest to simply use a 5532 for IC112/IC113, and then remove the diodes, resistors and transistors on the output of each opamp? so the output is directly connected to the opamp output?

Dimitree said:

thanks for the replies!

so you suggest to simply use a 5532 for IC112/IC113, and then remove the diodes, resistors and transistors on the output of each opamp? so the output is directly connected to the opamp output?

Click to expand...

You should probably add a 50-100 ohm series resistor between 5532 output and jack, but yes it could drive typical loads.

JR

Jürgen Haible had a Dim D clone:

http://jhaible.com/legacy/subtle_chorus/jh_subtle_chorus.html

He used TL072 and added 620R resistors in series.

living sounds said:

Jürgen Haible had a Dim D clone:

http://jhaible.com/legacy/subtle_chorus/jh_subtle_chorus.html

He used TL072 and added 620R resistors in series.

Click to expand...

TL07x can only drive 2k loads (including feedback R)  so 620 ohms may be partially to lighten the load on the TL07x...  but that would work as most inputs are 10k+.

5532 is just better as an output stage IMO (or the original design with transistors).

JR

JohnRoberts said:

5532 is just better as an output stage IMO (or the original design with transistors).

JR

Click to expand...

Yes, I think in my Haible Dim D Clone I replaced several TL072 with NE5532. 

I didn't know about the JHaible clone,
I'm checking it, and some doubts came to my mind about the ouput section..
Maybe on the original schematic (linked on my first post) there is a mistake, indeed I can't undestand the purpose of R167 (22K), and looks like it's not used by JHaible.
Also, what's the purpose of the fet Q112 before the output section..I really don't understand what it is switching and how many signals are mixed at the output.

Dimitree said:

I didn't know about the JHaible clone,
I'm checking it, and some doubts came to my mind about the ouput section..
Maybe on the original schematic (linked on my first post) there is a mistake, indeed I can't undestand the purpose of R167 (22K), and looks like it's not used by JHaible.
Also, what's the purpose of the fet Q112 before the output section..I really don't understand what it is switching and how many signals are mixed at the output.

Click to expand...

The JFET is most likely effect in/out

R167 is probably a gain tweak to hold level similar between in/out

JR

Thanks again
Can I ask what’s the purpose of Q9 and Q10, located bewteen the modulation output and the BBD clock section? To reduce the depth of the modulation?

Also, Q111 seems to change the gain of the opamp IC111 before the output section, and is switched in mode 4. Do you know what’s the purpose of that?

Dimitree said:

Thanks again
Can I ask what’s the purpose of Q9 and Q10, located bewteen the modulation output and the BBD clock section? To reduce the depth of the modulation?

Click to expand...

Yes. See at the bottom of page 4 how the sweep range is changed.

Also, Q111 seems to change the gain of the opamp IC111 before the output section, and is switched in mode 4. Do you know what’s the purpose of that?

Click to expand...

The user's manual does not describe the actual differences between modes, but mode 4 is probably a more subtle one, requiring less depth;

JohnRoberts said:

The JFET is most likely effect in/out

R167 is probably a gain tweak to hold level similar between in/out

JR

Click to expand...

but with R167 in that place, then the wet signal is always connected to the output, then how can you bypass the effect only with the JFET?

Bumping this thread... (first post in years!)

R167 indeed is there to even out the relative gain between wet/dry.  There is no feed through to the output when that JFET is off.  You can sim this and see.  Essentially the output of the IC112 buffer will sink the feedthrough current.

My question about the original circuit is:  what the heck is the Q105 path for?  It picks off the dry input, low passes it at 200hz, and then provides a -14dB attenuated signal to the summing node.  I thought this may be a phase thing but my simulation doesn't really show that. 

Anyone know?

soultek said:

My question about the original circuit is:  what the heck is the Q105 path for?  It picks off the dry input, low passes it at 200hz, and then provides a -14dB attenuated signal to the summing node.  I thought this may be a phase thing but my simulation doesn't really show that. 

Anyone know?

Click to expand...

It looks just like a subtle bass boost.

That's valid.  Quite subtle though 1-2dB.  I suppose it's a relatively cheap way to get there so maybe it makes sense.  Perhaps it's needed to round out the tone of the dry + wet signal.

What's interesting is that some interesting things can happen by phase shifting bass and it looks like this circuit would do that, but it doesn't seem to. 

I notice the same circuit persisted in the Dimension C as well, so it was important enough to carry-over to that product as well.

abbey road d enfer said:

It looks just like a subtle bass boost.

Click to expand...

Maybe it helps mono compatibility? One characteristic I really liked about the Dimension D was that it widened signals but still worked in mono

soultek said:

What's interesting is that some interesting things can happen by phase shifting bass and it looks like this circuit would do that, but it doesn't seem to. 

Click to expand...

Phase-shift effects from a 1st-order filter are well known and are limited. If polarity was reversed, the boost would become a cut of similar amplitude.