Boss DS-1 (MIJ) build - transistor boost config over selecting transistors?

[Bucksears]

I recently got my prototype DIY build of a MIJ DS-1 going, and noticed two things:

1) It is a LOT thicker than any post-1994 DS-1 I've owned. So much so that I'm looking at reducing one of the coupling caps to make it a bit leaner.

2) Yes, that Q2 transistor boost is SLAMMING the preamp. I've tried selecting transistors with lower HFE (I have all three in the 250-270 range), but I can't keep doing that. After researching the Q2 boost, it seems like the only other option is to change the values of R6/R9 to lower the boosting.

I'm using an ECG1464 (7pin preamp) for the IC, 2SC732TM-GR & 2SC2240-GR transistors and 1S1588 diodes. I removed the JFET switching section and the diode/resistor for the ACA adapter, but left the input/output buffers.

Question: It seems that 'GR-level' transistors (i.e. in the 200-400 hfe range) are standard for this and the SD-1. Is modifying R6/R9 pretty much the best way to adjust the DS-1's Q2 boost?

I'll see if I can post pics of my layout tonight.
thx

[ElectricDruid]

Providing a link to the schematic you're talking about would be a courtesy to your readers, to prevent them having to go searching about on the internet looking for something that might turn out to not be the same as the one you're looking at.

[aion]

hFE is more or less an abstract figure when it comes to gain. The gain of the amplifier stage as a whole is set by the bias resistors and is significantly less than the transistor's hFE. You could use a 2N3904 (~180) or a MPSA18 (~800) and still get more or less the same result.

With that said, definitely focus on the bias resistors if you want to lower the gain—but don't think of it as an alternative to selecting transistors with lower gain, because that's not actually doing much, at least until you get down to an hFE figure where the transistor can't keep up with what's being asked of it.

This also means that if you're using the bias resistors from the original DS-1 circuit, then you're getting around the same level of amplification from this stage that the original DS-1 has. This is not to everyone's liking, and people have had good luck taking a cue from the Big Muff (which this stage seems to have been inspired by) and adding a volume control immediately afterward to act as a pre-gain control. The BYOC Orange Distortion has this mod baked in. But just be aware that it's part of the sound of the stock circuit, not a fault in your build!

[Bucksears]

Here's the schematic I used:



It varies slightly from the DS-1 Service Notes I also have, as the service notes have a C24 150pF cap to ground coming off of pin 1 of the IC, and R21 (on the emitter of Q3) is 10K. In the above schematic, C24 does not exist, and R21 is 100K.

I bumped up R9 to 47R last night, and dropped C9 down to .068uF, which is the value in the post-1994 schematic. That is getting me closer to what I'm looking for. I'm fully aware that changing these values is 'not an actual MIJ DS-1', but I was just using that as a baseline for tweaking.

[antonis]

R9 value and particular transistor V_BE set Collector current..
Emitter voltage level is very low (about 100mV, give or take..) so a V_BE upwards variation result into same Collector current for R9=47R..
(with some help of feedback bias configuration, of course..)

[cctsim]

You can also increase C4 to 470pF, 560pF or even 1nF reduce the gain without having to worry about the changing the bias.

If you want more gain, you can replace Q2 with a Mosfet like 2N7000 and change R6/R7 to 1Meg. This effectively replaces the Q2 with a super hard on booster (ZVEX).

[antonis]

You can also increase C4 to 470pF, 560pF or even 1nF reduce the gain without having to worry about the changing the bias.

Only for high frequencies..

If you want more gain, you can replace Q2 with a Mosfet like 2N7000 and change R6/R7 to 1Meg. This effectively replaces the Q2 with a super hard on booster (ZVEX).

I don't think Mosfets exhibit transconductance higher that BJTs..

[Rob Strand]

t varies slightly from the DS-1 Service Notes I also have, as the service notes have a C24 150pF cap to ground coming off of pin 1 of the IC, and R21 (on the emitter of Q3) is 10K. In the above schematic, C24 does not exist, and R21 is 100K.

I bumped up R9 to 47R last night, and dropped C9 down to .068uF, which is the value in the post-1994 schematic. That is getting me closer to what I'm looking for. I'm fully aware that changing these values is 'not an actual MIJ DS-1', but I was just using that as a baseline for tweaking.

I guess it boils down to making one sound like the other or not.

IIRC there's more than two variants of the DS-1, it takes some effect weeding it all out.  I haven't looked at the details for years.  (Boss CS-1 has similar more than expected variants.)

You have already covered the main points.  Not sure you need to change 22R to 47R.  Gain pot setting accuracy and gain pot tolerance could easily swamp that.


Notice a couple of subtlies like the JFET switch resistance and the C2+C3 combination.
EDIT: fixed my dumb typo on pic

[cctsim]

You can also increase C4 to 470pF, 560pF or even 1nF reduce the gain without having to worry about the changing the bias.

Only for high frequencies..


You can see at least a 6dB reduction in gain for the 1nF cap after 1kHz