Boss NS-2 adding highs and overdrive

[Max9999]

Hi, I have a Boss NS-2 here that adds highs and overdrive when the effect is engaged. The part on the schematic that gets switched in then is between the return and the output, with IC1.
Looking at the datasheet I got interested in this part of the circuit, it looks to me like some kind of pre-emphasis:




In LT-Spice I get this response:




If this is a pre-emphasis, where between the return and output would be the de-emphasis?

[duck_arse]

I'm no expert. you are nearly there - R19 and C11 will provide some boosted highs. follow across to R16 and C10, the values work out the same frequency cut. the cap across R21 rolls off the high freq gain, for stability.

[Max9999]

Thanks duck_arse, I have looked into R19 ( typo, R16!)  and C10, but in simulation they dont give the expected result, see below. There starts to be some filtering when I add series resistance before R15, but I expect the output of the VCA to be low impedance, right?

What am I missing?

[ElectricDruid]

I agree with Duck that R19/C11 are the pre-emphasis, and C16 is just a stability cap for that op-amp. He's also right to notice that R16/C10 give the same frequency, which is very suggestive and unlikely to be a fluke!

As to why your sim of the situation doesn't give the expected result, I'm not sure.

What am I missing?

Checking the datasheet, it looks like that VCA has a current output. The example application circuit on page 6 shows it followed by an I-to-V convertor. This isn't much of a surprise - it's an OTA, isn't it?

https://datasheetspdf.com/pdf-file/869412/Mitsubishi/M5207L01/1

That might make a difference?

[antonis]

R19/C11 form a HPF effectivelly set in parallel with R20 (corner frequency at about 7.2kHz), the equivalent impedance of which being the denominator in U2b Gain formula..
(neglecting C9 and Q4 output impedance..)

>There starts to be some filtering when I add series resistance before R15<

Instead of adding resistance "somewhere" in the bias path (Q9 is a humble Emitter follower, already bias lower than its "ideal" point,,), try to "extract" resistance from the signal shunt path (R16 + C10)..
(I'd leave R16 as it is and increase C10 value - in the mean of more highs "leaking" out of Q9 Base..)

[Mark Hammer]

1)  The NS-2 employs a more complex, and presumably "better" rectifier circuit than many simpler gates usually do, so I can't automatically see it as a problem.  But, duty requires me to note that envelope ripple can often be perceived as "distortion" in gates, compressors, limiters, as the gain is modulated at a very fast rate, if subtly so.

2) The specs for the M5207 indicate it is able to comfortably handle 2.3Vrms input.  Op-amp stage IC2b provides for unity gain of the full spectrum and R19/C11 provide for greater gain for content above 7.2khz.  This is complemented by R16/C10, which bleed off high frequencies from the output at the same corner frequency.  Specs also indicate typically very low output noise (60uVrms; that's 1/17th of a millivolt).  So, it is hard to see where there is much need to "protect" against either distortion or accumulated hiss.

All of that said, is the thread/OP seeking a "cure" for distortion and an unwanted treble boost?

[Max9999]

Whoops, I made a type in my second post.
I meant R16 and C10 don't give the expected de-emphasis result.

How is this a filter?

[antonis]

I meant R16 and C10 don't give the expected de-emphasis result.
How is this a filter?

Together with OUT2 impedance it's formed a low-pass filter on Q9 Base..
(actually a voltage divider with part of its lower part capacitive..)

[Mark Hammer]

Cast your glance over to R12/C7/Q5.  This provides additional filtering of high end, depending on which mode one is in.  I'm going a little cross-eyed trying to follow the schematic, but suffice to say that when Q5 gets turned on, it's drain-source resistance drops considerable, although I don't know to what.  But let's say, for argument's sake, that it drops down to 100R.  In tandem with R12, that amounts to a combined parallel resistance of just under 99R.  Along with C7, that provides a 6db/oct rolloff starting around 1.6khz. 

So, question: is the HF noise and "overdrive" quality contingent on which mode you've selected?

[Max9999]

Cool, so the output of IC1 is a current source that makes the filter work.
I see now how this SHOULD correct the emphasis filter.

I threw the thing in a true bypass looper in the effects loop of an amp. Whats weird is that the bypass with the effect disengaged is a bit lower in output ( as expected with the source followers inside), but the output with the effect engaged is a bit higher.

This leads me to believe the difference is somewhere before or after the VCA. Can it simply be some mismatched resistors in the pre and de-emphasis? Maybe 5% high 5% low?

Also my model has a trimpot inside that is not shown on the schematic. Boss is unwilling to provide me a schematic due to company policy. Internet peeps are talking about this trimpot being there to balance the effected with the unaffected output, but I feel reluctant to just go and turn it.

So there are two options for the slight increase in distortion I think: the pedal is distorting a bit with the 20dB gain at IC2B, or the amps 12ax7 effects loop return has higher thd because of more volume into it.

Mark Hammer I do not hear a significant change between modes.

I think I should trace the trimpots function or experiment with disabling the pre and de-emphasis.

Thanks everybody.

[ElectricDruid]

Before you start unsoldering parts from the circuit, you should definitely tweak the trim pot. It's simpler, and reversible. You can always put it back where you found it - before you move it, get a fine-line permanent pen, and make a line off the part that moves onto the part that doesn't so you can see where it was. Then tweak away and see what happens and if you can figure out what it's doing.

[Max9999]

It turns out that my version actually uses THAT2181. The trimpot ( with two legs, never saw that), is connected to pin 3 in the dil package. This is the negative sens of the voltage control which controls gain, see text of datasheet attached.

I feel confident to turn that thing now, the isp decimator uses a similar sceme.

So the datasheet sais that when I measure the same voltage at pin 3 as pin 2 the gain is 1. Is this dc? Ac? Ofc I am going to tune it by ear, but what if I want to be precise and measure it with a dmm ?

[antonis]

DC..!!

In your circuit, pin 3 sits at 4.5v so, for unity current gain, pin 2 should also sit at 4.5V..
Actually you should read exactly the same voltage at pins 2 & 3 using the same DMM, which voltage might deviate from "ideal" due to meter's loading..

You see, single supply circuits implement half-supply point for various purposes (bias, reference point, etc) hence IC1 "GND" (pin 1) is connected to +4.5V..

[Max9999]

It worked! Problem solved! Party time! Excellent!

The trimpot controls the gain of the THAT. It actually has quite some range, you could use this as a clean boost or attenuator.

Sorry for the infantile questions sometimes, it's maybe a soothing thought that I am eager to learn.