• TalkBass has been independent since 1998. Add your voice.
    Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
    Join freeLog in
    Want zero display ads or expanded classifieds tools? Compare plans.

a different kind of low pass filter

Nicely done! Who'd have thought you could get four gang reverse audio taper pots? I don't see any DC bias on the positive input of the first opamp however. Wouldn't you want to tie that 1k resistor on the left directly to the transistor emitter rather than through a 1uF cap and maybe bias its base a bit above half supply? The two 6.8nF caps could go to earth/ground rather than Vref, I suspect you started with a Sallen-Key high pass where those two caps were resistors and so would have provided the positive input bias when tied to Vref. I should have run my circuit at unity gain. I thought there was going to be an advantage to running at higher gain but that turned out to be wrong so I backed it down to almost unity gain, no idea why I stopped at "almost"! Someone who feels two poles is enough but wants better battery life could use a single opamp and your input buffer, or use a dual opamp with the other half as the input buffer, assuming someone sells two gang reverse audio taper pots. Normal audio taper pots could be used but they would rotate the wrong way.
 
  • Like
Reactions: Antisyzygy
Who'd have thought you could get four gang reverse audio taper pots?
I know. Good old ebay. You can get anything. And it's really the star of the show with this circuit. And those look to be fairly nice quality and very cheap.

TBH, I wouldn't normally use such an obscure pot in a design. As a touring muso, I want pots, jacks and switches I can replace easily. Plus a pot like this is a bit of a pain to install if your bass doesn't have a pickguard. These regular electronics pots only usually have 6mm of thread. In other designs I've used VCR's so I could use a "normal" guitar pot. I've also been looking into digipot ICs for these sort of things. But I haven't done any programming for many many years...

I don't see any DC bias on the positive input of the first opamp however.
Oops, you're right. I havent' had a chance to spice this, I was just free-styling from an on-line sallen-key calculator. I suspect it'd work anyway because the total negative feedback ties the inverting input to 1/2 Vcc and the fact that no real opamp has an infinite impedence between + and - inputs. However, you are right, this is poor design. Yep rebiasing the front end and ditching that cap would work. I guess you could also just tie that +input to the vref with a 1M resistor.

I suspect you started with a Sallen-Key high pass
I'd like to think it was a mistake like that, but really it was just late at night and I didn't think it through properly. :(

assuming someone sells two gang reverse audio taper pots
I havent' checked, but I'd think ebay from china will do the trick. But I would think you'd also be able to get dual gang 50KC's fairly easily as they are used in mixers and a few amps as sweep controls for mid-range eq.
 
Last edited:
The revised edition!

Active LPF schem revised.JPG


KHutch - right on mate. extra bias resistor at the transistor, and the parts count and complexity is reduced. And the emiter voltage stays close to half Vcc with the battery down to 6volts.

The filter seems to work fine...

Fully anti-clockwise:
Active LPF spice 1.JPG



Middle: (This is with the pot at 15K. A 100K audio taper will usually be around 15-20K at its midpoint.)
Active LPF spice 2.JPG



and fully clockwise:
Active LPF spice 3.JPG


Was thinking I'll probably never get 'round to ordering that crazy pot and actually making one, but now I think about it, I have a fretless jazz that this might be perfect for... :cool:
 
Last edited:
  • Like
Reactions: bdplaid
There are various options for outboard filter units. A floor stomper (multi) that allows various presets might work, or something that could mount on a strap. It might be easier than trying to rig up an active circuit in your bass.
 
The revised edition!

View attachment 600251

KHutch - right on mate. extra bias resistor at the transistor, and the parts count and complexity is reduced. And the emiter voltage stays close to half Vcc with the battery down to 6volts.

The filter seems to work fine...

Fully anti-clockwise:
View attachment 600252


Middle: (This is with the pot at 15K. A 100K audio taper will usually be around 15-20K at its midpoint.)
View attachment 600253


and fully clockwise:
View attachment 600254

Was thinking I'll probably never get 'round to ordering that crazy pot and actually making one, but now I think about it, I have a fretless jazz that this might be perfect for... :cool:
If any of you guys might want consider building this for me, I have a smoldering checkbook...
 
Um, well, yeah I guess I could build one, but it'll take 2-3 weeks for one of those 4-gang pots to arrive from asia, then there's a good chance it'll take me a week or so to find the time (although it's a very quick build of course), and then probably another couple of weeks to ship to the US from here in sunny Melbourne...

I guess I could make the circuit itself, and you could order the pot yourself, then have a tech wire it at your end...

Do you have room for a battery? And a pot that will be about 1 1/4 inch deep? Plus a circuit that'll be about an inch square and 1/2inch deep? You'll need a stereo jack as well...
 
Some time ago I built a variable LPF filter (shelf) for a project bass and was completely underwhelmed by it. To my ears, admittedly shot with years of standing next to loud drummers, it had little advantage over a simple tone control. (But as always YMMV.)

I ended up building a MM 2 band EQ preamp and it is still in there.
 
Um, well, yeah I guess I could build one, but it'll take 2-3 weeks for one of those 4-gang pots to arrive from asia, then there's a good chance it'll take me a week or so to find the time (although it's a very quick build of course), and then probably another couple of weeks to ship to the US from here in sunny Melbourne...

I guess I could make the circuit itself, and you could order the pot yourself, then have a tech wire it at your end...

Do you have room for a battery? And a pot that will be about 1 1/4 inch deep? Plus a circuit that'll be about an inch square and 1/2inch deep? You'll need a stereo jack as well...
Let me check on the sizes - I have a double battery box separate from the cavity so that's covered. stereo jack no prob. This bass has been carved into so many times...
 
Some time ago I built a variable LPF filter (shelf) for a project bass and was completely underwhelmed by it. To my ears, admittedly shot with years of standing next to loud drummers, it had little advantage over a simple tone control. (But as always YMMV.)

My experience with the switched cap tone filters matches this to a degree. The thing that distinguishes them from a simple tone control to my ears is the resonant hump just before the cutoff. So if I were going to do this I would want to have a "Q" or peaking control as we discussed above to give me a flat LPF to simulate a traditional tone control as well as the peaked LPF to get more bite or bark or whatever you prefer to call it, keeping in mind that some people call it that horrid sound that I hate! But having two (or more if you are willing to do more exotic switching) modes would make this circuit a good replacement for a traditional tone control in active instruments. The state variable designs that this thread started on do have a variable Q control so they would be fine choices too, if you don't mind using a quad op amp. We've been talking about Sallen-Key filters lately to save current with a dual opamp and to meet the design challenge of doing it with a dual but the infinitely variable Q of the SV filters could be an advantage to some ears.
 
  • Like
Reactions: bdplaid
I agree. and that's why I put that little 4dB resonance there. I think with the 4-gang thing it looks like we've found the op's circuit, but I sat down yesterday and designed a non-cascaded sallen-key (2nd order) lpf with a switchable resonance. the peak can be 4dB or 9dB. it's an incredibly simple circuit compared to a typical state variable implementation. and of course only uses the dual opamp and a handful of parts.

the original q filter was best with those low z pickups. the resonance of the pickup was a really high frequency so the filter swept through a more linear region of pickup response. this way the q filter could mimic the response of just about every pickup around.

if you use a q filter with traditional pickups that are essentially wound for passive use, you'll be sweeping the filter around in a very non-linear region of pickup response. in fact you'll be all around the pickup's actual resonant peak and with the addition of the filter's resonance you'll get a pretty huge peak quacking from your speakers each time you hit that frequency.

the idea with the switchable q is that you can switch to low q if you are around the pickup's resonance. well that's the idea anyway...
 
  • Like
Reactions: bdplaid
MEC (whoever they are, cannot find them directly only a couple of dealers in a web search) does make a dual, reverse audio, 250k pot with a push pull switch also available from here. They aren't cheap, at that price a pot would pretty much have to cook breakfast for me! Bourns claims to make reverse audio pots in their "pro audio" line but I can't find a dual in stock at any distributor. They do make Invalid Link Removed and Invalid Link Removed so I suppose one could learn to live with a tone control that works backwards. If you want to have two levels of peaking a dual pot with a DPDT switch would let you select the peaking as well as the frequency which is pretty cool. I keep looking at dual pots because I think two poles is a steep enough filter, really.

Now, I think if you were to disassemble a dual audio taper pot and swap the two wafers you would convert it to a dual reverse audio pot but taking apart a dual pot with a switch and putting it back together could be a bit of a nightmare I suppose. I guess I could buy a couple of the Bourns pots and try to modify one and still be in for only half the price of the MEC....
 
  • Like
Reactions: bdplaid
... When I look at ToneStyler's curves I see a critically damped response unlike the one you get when you just switch in one of several different capacitor values without any pot connected. And indeed in some of their product photos I see a resistor partially hidden behind the switch. So evidently they intentionally damp their circuit with a series or parallel resistance to eliminate the mid band peak that you get without a damping resistor. Some of us like that peak and the "Alembic" state variable filters can produce it too, but many do not like that peak. In my experiments a parallel resistance is more effective than a series resistance in damping the peak and below I show some curves for a switched capacitor tone control (this one really is stepped) with a resistor in parallel with each of the six capacitors. Every capacitor value wants a different resistor to get roughly the same damping but since you are already switching the caps, the switched resistors in parallel just come along for the ride...

View attachment 598800

Hate to derail the thread...

Khutch, just a question... I want to do a passive tone control with a three-way switch for three different caps (roughly .1uF, .047uF, and 0.2uF) and a tone pot (likely 500k, so that fully open is a near-full bypass).

I like the resonant peak that occurs on a typical tone pot as it's dialed down. Should I put any resistors in this circuit? I initially thought not, but your graph does show some minor resonant peaks and you say there was a resistor in parallel with each of the six capacitors.

Thanks!
 
MEC (whoever they are,

Invalid Link Removed.

52 euros for a pot that's hilarious! Not including postage too I'll bet. I've bought whole basses for less than that! Yep the MEC stuff is nice quality, but really crazy in price. It's nice to have a resonance switch at the tone pot itself, but for that money, I'd put it on a push-pull at the volume or somewhere else. Or just drill for a mini toggle...

If you have the patience for postage, then really the ebay stuff out of china/HK is fine IME and very reasonably priced.
Active LPF 2nd order.JPG


A ridiculously simple circuit that would I think not be a million miles from the response of a q filter. I'd be a little nervouse about the switch popping though, so perhaps C4 should switch to 4.5v. Virtual earth instead of actual earth. THis is much more like a circuit that I'd try in one of my own basses. Plus I have some C50K's kicking around.

The sweep is still from 100 to 10K.

Any low current and fet input opamps will be fine. The LF 442 is an obvious choice, but there's lots of others. TL072, LF351, TL062...

I guess if you were to use low Z pickups, you'd need a little gain, so the first opamp could be redesigned with gain set resistors. Or even adjustable gain. And the Q switch could be a centre off toggle so with a couple of extra caps would have three Q settings.
 
Last edited:
thought I'd resurrect this thread because I built this circuit the other day and put it in my fretless jazz.

if there's still interest, I'll post more details- I changed the circuit a little, plus added an active pickup mixer to the front end. plus I could record some samples I guess...