Double Bass EQ question

[Frank77]

A few questions for the scientists/engineer/technicians among us:

Most commercially available 7-band EQ pedals for bass guitar seem to have center frequencies at 50, 120, 400, 500, 800, 4.5k and 10kHz. If I wanted to build a 7-band EQ specifically for an upright bass, would these center frequencies still be a good choice? Considering most notes on the double bass are in the 40-350 Hz range, do the 4.5k and 10k stages have much of an impact, or does the instrument actually produce harmonics that high?

Any input/suggestions welcome...

 

[GrowlerBox]

A few questions for the scientists/engineer/technicians among us:

Most commercially available 7-band EQ pedals for bass guitar seem to have center frequencies at 50, 120, 400, 500, 800, 4.5k and 10kHz. If I wanted to build a 7-band EQ specifically for an upright bass, would these center frequencies still be a good choice? Considering most notes on the double bass are in the 40-350 Hz range, do the 4.5k and 10k stages have much of an impact, or does the instrument actually produce harmonics that high?

Any input/suggestions welcome...

Speaking as none of the invited participants, I'll gatecrash your party anyway. If I were building (or commissioning) an EQ for double bass, it would be parametric rather than graphic. There is much greater variation between double basses, their strings and setups, than I feel exists between bass guitars, and the problems of amplifying a large resonant box which is sensitive to its environment make fixed-band equalisers less useful in general. Which isn't to say that they're useless, but if you're going to all that trouble ...

 

[Frank77]

Hey, everyone's invited...I should have phrased my opening statement differently I suppose.

I am not excluding the possibility of going the parametric route, but I am still on the fence...

I have a semi-parametric EQ (fixed "low" band, adjustable mid and high) on my old, glitchy and probably-about-to-die Boss ME-8B. I do get good results with it, but I find it takes me a considerable amount of tweaking time to get there...the graphic EQ is more intuitive (at least for me) and it's been done to death: there's tons of schematics I can find on the internet to get started.

On the other hand, I could probably put together a more sophisticated device similar to (but certainly less refined than) this: [Invalid or Expired Link Removed] for, I don't know, maybe 50$ or even less, but honestly there's so many adjustable parameters that I think I'd be somewhat intimidated and could very well get lost trying to dial in a decent tone.

 

[f.c.geil]

While the fundamental pitches of a double bass fall between 40 and 350 Hz (same as bass guitar, incidentally), you have to consider the harmonic frequencies as well. The really high frequencies (8K and up) add "air" and "space" and "life" to your instrument. With that being said, the strings being used on a double bass are not particularly bright, so I would probably only consider the first and maybe even second harmonics. So, if I were building an EQ pedal, I would probably center frequencies at 50Hz, 150Hz, 300Hz, 500Hz, 1.2kHz, 2.4kHz, and 10kHz.

 

[Jason Sypher]

I've got one of these for sale in the classified. Buy one for me or from someone else. It's a very powerful and versatile option. Cheers.

[Invalid or Expired Link Removed]

 

[Frank77]

Thanks for your replies.

I've got one of these for sale in the classified. Buy one for me or from someone else. It's a very powerful and versatile option. Cheers.

[Invalid or Expired Link Removed]

I've heard good things about the Headway device, but I'm not really looking to buy anything ready-made. I'm currently studying electronics engineering (as part of my "I'm no longer in the army" plan) and have access to all the tools and test equipment to make one of my own on the cheap. I don't really mind the number of hours I'll put in as I see this as a learning opportunity / fun project to do anyway. Plus, I don't really need the device in that much of a hurry: I have a gig in February but the glitchy Boss ME-8B still works. Kinda.

A question for those who are more partial to para EQs: What procedure or drill do you go through to hunt down the right frequency to cut/boost with the least amount of time? I'll play around with the one that's on the Boss and if I become friends with it I'll probably go the parametric route...

Cheers!

 

[longfinger]

If you are going to build an EQ pedal, it will be useful to also include a variable set point high pass filter in it. Have a very steep and hard fixed roll off for frequencies below 25Hz, and a user-movable more gently roll off for frequencies of say 25Hz up to 125Hz or so. Also include an input that works nice with piezo pup. (input impedance of at least 1MOhm, to 10MOhms).

 

[Adagio]

A question for those who are more partial to para EQs: What procedure or drill do you go through to hunt down the right frequency to cut/boost with the least amount of time?

1 - Listen to determine what is the approximate frequency of the feedback.

2 - Turn the corresponding EQ knob until that horrible sound stops.

3 - Repeat step 1 until you hear at least the drummer.

 

[DoubleMIDI]

If your sound is bad, dial in a medium wide Q/bandwidth, boost a lot maybe up to max, turn the frequency until it sounds worst.
Then cut the frequency until t sounds more or less OK, vary Q/bandwidth and also the cut until it sounds best.

There might be more than one boosted frequency, so having to full parametric equalizers is good.

For the high impedance piezo buffer and the high pass (low cut) filter have a look at Francis Deck's HPFpre and his Quick'n'dirty schematics. You might want to add a parametric or graphic equalizer to it. Maybe also a second channel to mix a microphone in.

 

[Frank77]

I guess I should give more details on my actual setup...

The most important reason why I want to do this is because on most gigs, I double on the bass guitar and I use the same amp for both instruments...so what I'd like to achieve is something that allows me to keep the same EQ and volume settings on the amp no matter which instrument I use.

When playing BG, I leave my amp's EQ settings at noon and dial my tone with the electronics of my L2500. When I play DB, I usually cut 9 dB of bass and mid, and 6 dB of high mids on my amp's EQ. I use a Hpf-Pre 2 between my Realist-equipped DB and my amp, with the filter usually set at about 1-2 o'clock.

So my current setup looks like this:

DB--->Hpf-Pre--->Input "A" of A/B switch--->Amp

L2500------------->Input "B" of A/B switch--->Amp

Right now, I could achieve what I'm looking for by putting the ME-8B between the A/B switch and the amp, and create two separate patches with EQ and volume settings for each instrument. However, as I mentionned, the ME-8B has become unreliable and I'm not too crazy about its effects anymore (most of which I barely ever used anyway). I'll probably decide to go with dedicated pedals for each of the few effects that I do use (compression, fuzz, octaver, and the very occasionnal t-wah) in the near future.

Does this make any sense or should I take it easy on the crazy pills?

 

[f.c.geil]

You could try using an Empress ParaEQ or similar. Since it has a boost function, it will allow you to balance output. I assume you wish to keep the character of the two instruments intact, allowing them to sound different, but using the same EQ setting to achieve satisfactory sound. I would think something like the ParaEQ would do that for you. If, on the other hand, you are trying to make both instruments sound the same, I'd first ask "why?" but then would be at a loss for a product that would do that. The only solution I would have for that situation would be to have two EQ pedals, one on each instrument cable, perhaps with a boost on one of them to equalize the volume, then into an A/B switch into your amp.

 

[Frank77]

I assume you wish to keep the character of the two instruments intact, allowing them to sound different, but using the same EQ setting to achieve satisfactory sound.

Correct. In other words, I want to make the two output levels similar (obviously my BG's active pickups are louder than the piezo on my DB) and some EQ adjustments to one instrument without changing the settings on my amp. It does seem like a device similar to the paraEQ would do the trick nicely.

 

[longfinger]

A two channel amp would solve this issue pretty simply.

But that won't take away the itch to build a neat gadget! ;-)

 

[Frank77]

A two channel amp would solve this issue pretty simply.

But that won't take away the itch to build a neat gadget! ;-)

There's a bit of that...and also the fact that right now I don't gig enough to justify the investment. I admit this solution would be ideal, though. Maybe someday...(perhaps a "from me to me" graduation gift in 2 years).

 

[Champagne]

Speaking as none of the invited participants, I'll gatecrash your party anyway. If I were building (or commissioning) an EQ for double bass, it would be parametric rather than graphic. There is much greater variation between double basses, their strings and setups, than I feel exists between bass guitars, and the problems of amplifying a large resonant box which is sensitive to its environment make fixed-band equalisers less useful in general. Which isn't to say that they're useless, but if you're going to all that trouble ...

+100

Graphic eqs impart more phase distortion than parametric. Instead of creating a large bell with one band with a parametric, you will use 5 narrow bells (bands) on a graphic to create the same with each imparting some distortion to the signal.

 

[Champagne]

If your sound is bad, dial in a medium wide Q/bandwidth, boost a lot maybe up to max, turn the frequency until it sounds worst.
Then cut the frequency until t sounds more or less OK, vary Q/bandwidth and also the cut until it sounds best.

There might be more than one boosted frequency, so having to full parametric equalizers is good.

To add to this with a engineer's spin, it all depends on what you are trying to filter. "What is the bad sound?"

Some instruments have imbalances as notes are played, one note may sound really loud compared to others. For instance, your C is over resonant so you would sit a narrow bell (notch filter) at 65.41HZ and cut it a bit. You can also use a boost notch and sweep around for the offending frequency then cut.

Boosting a wide band to the max is very dangerous and misleading. If you do a wide boost, especially to lower frequencies, you are going to overload the amp and the sound will change as a result due to taxing the preamp and amplifier. You can also cause a amp's protective limiter to lock and load on your sound and destroy it. It will sound bad no matter what. Trust me on this. You can also screw your ears up for a while listening to something piercing as you sweep around trying to find what is so bad.

Lastly, you are better off cutting more than boosting. That is the biggest problem for most people. The first thing they want to do is boost something. This leads to loss of headroom big time and it can imbalance your gain staging if there a few components in line with each other.

Learn to cut first then boost later. That will lead you to cleaner sound and a more balanced signal path. If you can boost and hear something bad, you can also cut and hear something good. It is all about training yourself to do the opposite. Once you do, you will have actually trained your ear to find the good qualities and bring them out and be able to set them in the pockets that compliment the other instruments in the arrangement.

 

[drurb]

+100

Graphic eqs impart more phase distortion than parametric.

This is not necessarily true at all. It depends greatly on the design of the filters and the boost/cut applied.

Lastly, you are better off cutting more than boosting. That is the biggest problem for most people. The first thing they want to do is boost something. This leads to loss of headroom big time and it can imbalance your gain staging if there a few components in line with each other.

Learn to cut first then boost later. That will lead you to cleaner sound and a more balanced signal path. If you can boost and hear something bad, you can also cut and hear something good. It is all about training yourself to do the opposite. Once you do, you will have actually trained your ear to find the good qualities and bring them out and be able to set them in the pockets that compliment the other instruments in the arrangement.

It is a myth that cutting is preferred to boosting and it is simply not true that if "you can boost and hear something bad, you can also cut and hear something good." Neither is the converse true. Whether it is more desirable to boost or cut is determined by what is the minimal adjustment that can be made to achieve the desired spectral profile. Here's an example. Suppose that the desired spectral profile is achieved by a 2 dB boost in a 1/3 octave band around 1 kHz. Well, your best bet is to provide that boost. Cutting all other available bands around that center frequency would be most undesirable as it would result in a rippled spectrum that would stem from the overlap regions between individual filters.

The "boost rather than cut" myth seems to have stemmed from the probability of boosting too much and overloading the input of the device in the signal chain that follows the tone-shaping circuit. This can be prevented with much equipment by adjusting the overall gain of the tone-shaping device itself.

Finally, a word about phase distortion. It simply hardly matters at all when it comes to the amplification of the double bass. In fact, it matters far less than people typically believe even if one is considering critical listening to amplified music. The actual scientific psychoacoustic literature bears this out.

As for "headroom," this might be helpful.

 

[Champagne]

This is not necessarily true at all. It depends greatly on the design of the filters and the boost/cut applied.

What would the majority be? Catch my drift.

It is a myth that cutting is preferred to boosting and it is simply not true that if "you can boost and hear something bad, you can also cut and hear something good." Neither is the converse true. Whether it is more desirable to boost or cut is determined by what is the minimal adjustment that can be made to achieve the desired spectral profile. Here's an example. Suppose that the desired spectral profile is achieved by a 2 dB boost in a 1/3 octave band around 1 kHz. Well, your best bet is to provide that boost. Cutting all other available bands around that center frequency would be most undesirable as it would result in a rippled spectrum that would stem from the overlap regions between individual filters.

HA! No kidding. That would be the way I would do it too. You see, you and I are kind of alike!

The "boost rather than cut" myth seems to have stemmed from the probability of boosting too much and overloading the input of the device in the signal chain that follows the tone-shaping circuit. This can be prevented with much equipment by adjusting the overall gain of the tone-shaping device itself.

That all depends on the circuit design and the limitations of it. A lot of amps sound horrid boosted all over the place regardless of the input and output gains and whatever other knobs are in between.

Finally, a word about phase distortion. It simply hardly matters at all when it comes to the amplification of the double bass. In fact, it matters far less than people typically believe even if one is considering critical listening to amplified music. The actual scientific psychoacoustic literature bears this out.

Well, I am coming from an engineering point of view and that is a valid point. Maybe it doesn't matter all that much after it is pumped through a power amp but I will say this: Almost every gigging band I record comes in with "their sound". Live sound is different than studio recording, cool. I pick the mics suited for their tone and throw the faders up and ask them if that is what they are after. I will even swap mics too if they don't like it. Bare in mind, these mics aren't always jammed down the speaker's throat the amps are in rooms not closets and sometimes sharing the air with the whole band. 9 times out of 10 they aren't totally into it, so what do I do? I pay a visit to their amps and what do I see?

:[Bass Amp]: Bass knob turned to 9 or 10 a built in graphic with boosts all over the place. What do I do? I back everything down on the eq, and work subtractive EQ. If their settings support it, I invert their whole eq carving to be primarily subtractive (example: big ass bass and treble boost into a mid frequency valley with a slight high frequency bell for definition). I bull **** you not: 95% of the time they are wowed and the majority of rigs are similar to this.

Anyway, All eqing is a taste thing. Do what ya like and how ya like it. Just know there are two sides to the EQ coin.


Headroom: [Invalid or Expired Link Removed])

 

[drurb]

I'm well aware of the "sides" of the EQ coin. In fact, I'd say that two sides doesn't even capture it. I'm coming from an engineering point of view and several other relevant professional points of view. By the way, there is no basis of which I'm aware to contend that "graphic eq" filters, in general, contribute more phase distortion than do parametrics. Again, it's a matter of design.

Your approach seems wise as you seem to attempt to use as little tone-shaping as possible. The desirability of that has virtually nothing, however, to do with phase distortion.

As for headroom, the link you posted is a dead end at Wikipedia. I assure you that the definition to which I linked is quite correct.

In any case, there doesn't seem to be much about which we disagree once the details are discussed. Your own generalizations don't coincide with what you necessarily do and that's a good thing.

 

[Frank77]

Ok...how does this sound:

A parametric EQ with 3 selectable frequencies (low, mid, high), each of course with their respective gain, but also each with a dedicated bypass switch (so you can avoid adding 3 stages of noise if you really only need to tweak the mids, for example), 3 position switches to choose different values of Q for each stage, and a master gain.