current boss pedal power filtering examination

[Rob Strand]

but the active filter still needs those electrolytics unless I'm mis understanding? Im assuming we're considering 100uf "big".

With RC filters and 100uF it's going to be electrolyolytics.  But with the active filter and 4.7uF or 10uF you can use larger package SMD ceramics or tantalums.

[marcelomd]

but the active filter still needs those electrolytics unless I'm mis understanding? Im assuming we're considering 100uf "big".

With RC filters and 100uF it's going to be electrolyolytics.  But with the active filter and 4.7uF or 10uF you can use larger package SMD ceramics or tantalums.

Looking at the schematics, Boss uses the active filter/capacitance multiplier PLUS one or two 100uF electrolytics.

For MY purposes - getting the size down - it more or less defeats the purpose.

[Rob Strand]

Looking at the schematics, Boss uses the active filter/capacitance multiplier PLUS one or two 100uF electrolytics.

For MY purposes - getting the size down - it more or less defeats the purpose.

That's not really an apples to apples comparison.   You don't *have* to have the 100uF capacitor with an active filter.  An active filter can provide a lot more ripple rejection than an RC filter even without the 100uF cap.

There's also the option to use a smaller output cap.

There are circuits where it's not all about ripple rejection.   There's circuits where output impedance at DC and at high frequencies needs to be considered.   If you have a transistor fuzz circuit it doesn't need a 100uF supply capacitance.    However if you have something switching on the rails, like an LFO, then the 100uF cap can help prevent supply noise propagating in the circuit.

Like many design problems there's many angles to finding the best solution.

[m_charles]

If anyone is interested, from the links Rob suggested, I found the most current trace done from a 2023 SD-1 waza. I looked up the diodes from the part numbers and these are the types.




Sorry for the misspelled "Shottkey", haha.

one thing that seems a bit confusing is why are the schottkey and the rectifier needed. both do the same thing here right? protect against reverse voltage?
Also, why the extra tabs on the DC jack? 4 and 5?

[m_charles]

another observation, why use an 11v zener to keep that voltage stable to the transistor? maybe just an error by the person tracing? supposed to be 9v?

[Rob Strand]

You can see the same things on the SMD version of the BD-2 in the link I posted earlier.  It's probably like that in all respects.

another observation, why use an 11v zener to keep that voltage stable to the transistor? maybe just an error by the person tracing? supposed to be 9v?

Traditionally Boss used 11V zeners for over voltage protection.  I think it just comes from that.   Not all a products use a zener there.

thing here right? protect against reverse voltage?

I mentioned that earlier on.   The Schottky is series protection, unlike parallel protection, which doesn't fry anything when you reverse the supply.   I think they tacked that onto the units and didn't remove the parallel protection.   So the rectifier is kind of a left over.

Also, why the extra tabs on the DC jack? 4 and 5?

No idea.  Could simple be that the person who drew the schematic used the wrong part.   However it could be related to Bosses supply chain. 

Sorry for the misspelled "Shottkey", haha.

one thing that seems a bit confusing is why are the schottkey and the rectifier needed. both do the same

I did that at work once in one of my documents and a work colleague was reading it out and made a point of pronouncing it as a shot-KEY diode.

[antonis]

Considering D2 ideal, Q1 configuration is, IMHO, just a capacitance multiplier..

That said, Q1 Emitter voltage is highly dependent on current draw..
(practically, no voltage regulation..)

[Rob Strand]

That said, Q1 Emitter voltage is highly dependent on current draw..
(practically, no voltage regulation..

In practice it varies very little current as the base resistor base resistor value is quite low.   It's not going to be far off the VBE drop for the transistor (maybe 0.7V).   The pedal current is fairly constant.

When the zener is 11V it's basically a filter.   The zener only kicks in during over voltage.

[duck_arse]

Also, why the extra tabs on the DC jack? 4 and 5?

see, this is why we argue about correct diagram drawing. lug 2 connects external supply to internals, and might come with reversed polarity. D1 will block reverse, hurrah. but will also drop some voltage. lugs 4 and 5 are switched contacts, such that when the external is in plugged, the switch is open and the battery is isolated from the circuit and external. when the external is instead out plugged, the switched contacts are short, and the battery is powering the effect. and so as to avoid that D1 voltage drop, pin 4 connects the other side of the diode. because you'd never put the battery in the clip backwards, see.

[m_charles]

Also, why the extra tabs on the DC jack? 4 and 5?

see, this is why we argue about correct diagram drawing. lug 2 connects external supply to internals, and might come with reversed polarity. D1 will block reverse, hurrah. but will also drop some voltage. lugs 4 and 5 are switched contacts, such that when the external is in plugged, the switch is open and the battery is isolated from the circuit and external. when the external is instead out plugged, the switched contacts are short, and the battery is powering the effect. and so as to avoid that D1 voltage drop, pin 4 connects the other side of the diode. because you'd never put the battery in the clip backwards, see.

I actually figured that out between last night and today, haha.

[m_charles]

Rob, as you mentioned, there were many variations on this. I'm wondering why on some of them they have the large cap to ground before and after the BJT? Maybe the cap to ground before the BJT on especially high gain circuits, although that still seem unneeded to me with my limited knowledge? seems like the first one is pretty unnecessary.
Images below show the Blues Driver 2009 had both, the SD-1 Waza only after. Again the caveat here is I'm assuming the person that traced it made no errors.

[Rob Strand]

I'm wondering why on some of them they have the large cap to ground before and after the BJT? Maybe the cap to ground before the BJT on especially high gain circuits, although that still seem unneeded to me with my limited knowledge? seems like the first one is pretty unnecessary.

On the BD-2 I suspect the input cap isn't doing much.   You could argue it might help remove ripple in marginal cases.   

For the output cap, it helps lower the impedance on the effect side of the supply.   Justifying a *need* for it is a fair amount of work and testing.  I suspect some amplifier circuits it might help to have some capacitance these, maybe 10uF would do it, but the 100uF probably isn't required.   

For other pedals which have strong varying currents, either of the 100uF caps can also help prevent polluting the input power.   The output side cap would prevent polluting its own supply rail; for example a pedal with an LFO.

As mentioned in one of the earlier posts.   It's not always about filtering.  The active filter is about filtering but it doesn't store energy like a cap.  Caps on the rails will provide a energy reservoir to smooth out current variations.

Boss have a recipe and in many cases they stick to it because they haven't had any issue.  Now and then someone might stick their neck and and change the recipe.

[antonis]

C11 without a low value resistor in series with IC1A is looking for trouble..
IMHO, C11 it's redundant 'cause R5//R4 & C2 LPF corner frequency is much lower than IC1A output impedance & C11 one..

[PRR]

C11 it's redundant

I don't know what that added opamp does. It is begging to oscillate and it adds hiss. All the bias loads are very high impedance, we don't need bias buffer.

[antonis]

I don't know what that added opamp does.

Maybe it serves for a Vref current hungry circuit feeding..
(although I do agree for very high impedance bias loads..)