Bass too loud in stage montior - any devices that can help?

[JCJR]

oky dokes - frequencies are 50hz (needed -12dB) and 125hz (needed -3dB) and this gave me a clear unmuddied lower end.

Hi miden. First will attempt a clear description of something that might work OK, and follow a few explanations in case anyone would be interested, or want to check the reasoning, or adapt to different assumptions.

This assumes you use a tip-sleeve "guitar style" cable from a mixer line out to an amp (or powered speaker amp) input. Other things could be figured out if neccessary.

You might try to find a quarter inch plug with a "unusually roomy" back shell. Or alternately if space is too cramped just solder in the two little parts to the jack and protect it with a few layers of tape or preferably heat-shrink tubing, and just be careful not to tear up the cable with rough handling.

You could take a cable you already have and dissect one of the plug ends, or snip off one end and use a new plug. The modified end should be marked and always plugged into the amp (or powered speaker amp) input.

One quarter-watt film resistor value of 1.2 kOhm. Eighth-watt would work as good but they get kinda tiny. Probably 5 percent tolerance resistor is good enough but you could use 2 or 1 percent if desired.

One capacitor value of 0.68 uF or 680 nF. Which is the same value. MicroFarad is 10 ^ -6 and NanoFarad is 10^-9. Less desirable capacitor types for this use include electrolytic, tantalum and ceramic. They would probably work but are less desirable. More desirable types would include polyester, mylar, mica. About anything that ain't electrolytic, tantalum or ceramic. If you find one in a junk box or cheap on a store shelf with real high voltage rating it will work fine but for this use any voltage rating bigger than 25 volts ought to be fine. 5 percent tolerance ought to be good. Probably not worth the money to hunt down a 1 percent cap. 10 percent tolerance would work but just be a little less "predictable".

Your cable has a center conductor and a shield wrap.
_1 You would wire the shield wrap to the plug SLEEVE as normal.
_2 You would solder the center conductor to one wire of the cap and then solder the other cap wire to the plug TIP.
_3 Solder one end of the resistor to the plug TIP and solder the other end of the resistor to the plug SLEEVE.

If you can find some kind of little vise or maybe even use some grip-lock pliers to hold the plug in a convenient attitude for soldering, you would probably first solder one end of the cap to the center conductor. Then you could poke both the other capacitor wire and one of the resistor wires into the plug TIP terminal and solder them both at the same time. Similarly poke both the shield and the other end of the resistor to the plug SLEEVE terminal so you can solder them both at the same operation. It is good to use little bits of tape or heat-shrink tubing around the center-to-cap-to-TIP terminal stretch to make it unlikely that those bare connections can short out to the SLEEVE terminal.

====

Explanations: Analog can be as accurate as one throws money at, but tends to be "in the ballpark" kinda thing. Using single off-the-shelf component values you can get "in the ballpark" of what you want. To get closer requires a lot more work and probably isn't justified in this example.

This 6 dB per octave hipass "ballpark behavior"-- It is "about -3 dB per octave" for about 2 octaves then it is "about -6 dB per octave" from then on. So if you want -12 dB at 50 Hz, you would get "about -6 dB" at 100 Hz and -3 dB at 200 Hz. Which is a little different from your above spec but might be purt close in practice. In simple circuits it can't be adjusted any finer so far as the shape of the curve.

You can only move the curve up and down by tuning the filter. For example you could decide to settle for -9 dB at 50 Hz in order to get a lower -3 dB frequency or whatever.

The -3 dB frequency (200 Hz) is called the "cutoff frequency" or "center frequency".

You want the impedance ratio between the mixer output and the filter as big as feasible, and you also want the impedance ratio between the filter and the amp input as big as feasible. Doing some googling, it appears that nowadays TYPICAL line outputs have 150 Ohm or less impedance, and several big-name pro power amps seem to use 10 kOhm unbalanced input impedance. Long long ago working musician mixer output impedances averaged a little higher than 150 Ohms with only REAL EXPENSIVE gear havin the real low output impedances. And long long ago I think the amp input impedance might have averaged somewhat higher than 10 kOhms.

So it doesn't matter much if your amp input impedance is higher than 10 kOhm but if your mixer output impedance is substantially bigger than 150 Ohms then maybe different parts values would work better.

We can put the filter impedance "in the middle" by taking the geometric mean of the typical input and output impedances: (150 * 10000) ^ 0.5 = 1.225 kOhm. In that case it turns out that a 1.225 kOhm filter would be about 8.2 X BIGGER than the 150 ohm mixer output impedance (and it doesn't matter if your mixer impedance is lower like 120 ohms or 50 ohms or whatever). Also the 1.225 kOhm filter is about 8.2 X SMALLER than a 10 kOhm amp input impedance (and it doesn't matter if your amp might be bigger like 20 k or 50 k or whatever). The bigger the ratio the better. Too-small ratios can make the filter curve get even sloppier than usual for a 6 dB per octave filter but too-big ratios no problemo.

They don't make resistors and caps in all possible values. There are sets of common easy to find values as listed here: http://www.kennethkuhn.com/students/rlc_values.pdf

If we were real picky we could solder together several parts to get closer but in this case it is probably not worth the trouble.

The closest standard resistor value is 1.2 kOhm.

This formula: 1 = 1 / (2 * PI * R * C * F), where R is resistance in Ohms, C is capacitance in Farads, and F is frequency in Hz.

If you know all but one of the parameters, simple algebra just lets you move the unknown term over to the left side and solve it. Because we know that F is 200 Hz and R = 1200 Ohms, we can find the capacitor value: C = 1 / (2 * PI * R * F) = 6.63 x 10^-7 Farads. Which is either 0.663 X 10^-6 (microFarads) or 663 X 10^-9 (nanoFarads).

According to the above-linked table of common capacitor values, the closest appears to be either 0.68 uF or 680 nF.

So assuming that we build the circuit with "exact perfect tolerance parts" of 1.2 kOhm resistor and 0.68 uF cap, what is the expected -3 dB point?
F = 1 / (2 * PI * R * C) = 195 Hz.

Which is surprisingly close to what we wanted! It will of course be "in the ballpark" of 195 Hz because the resistor and cap won't have EXACT values. But nowadays even 5 percent parts are sometimes surprisingly close to the nominal value. So this should be "about -6 dB" at 100 Hz and "about -12 dB" at 50 Hz.

Been a dog's age since I last did this calc so if anyone notices errors please point them out.

[miden]

Outstanding! thanks heaps. I am going to put all this into a little aluminium kit case, with 6.35mm jacks either side with the speaker one marked with paint pen and an engraved legend. Simply because I like "neat" hahaha!

I'll get on to that. Still deciding on whether to go with the iPad 3/4 to run the X Air 16, or a MSP 3/4.

[Winstontaneous]

Can you raise it off the floor (milk crate, amp stand, foam, blanket)?
Decoupling will reduce the boundary effect that emphasizes low frequencies.

[JCJR]

Hi miden. What is an msp 3/4?

I initially ran my xr18 with a several years old but was expensive and nice when new Motorola 10.5" Android. As best I recall it worked ok but after maybe a year the old thing finally went into permanent panic which hard reset couldn't fix.

So I put in an even more ancient win10 laptop. Originally was win7 with hard drive and hopeless slow even for web browsing but not maddeningly slow with maxed ram and an SSD and win10.

I used ethernet mainly so the laptop wouldn't get confused by my home network and all the neighbor's wireless networks. Hardwire Ethernet to the xr18 with WiFi entirely disabled on the old laptop. It runs nothing except the bejringer control software.

About the cheapest 10" pad might be Amazon Kindle fire because amazon uses em as loss leaders and the latest model screens are rather sucky compared to such as expensive samsung.

I have a $60 7 inch Kindle fire but never tested it with the behringer software. Eyes have been getting weak and even a 10 inch pad would want a magnifying glass for my eyes nowadays.

Maybe sometime should test my 7 inch Kindle fire just to know how well a 10" Kindle would be likely to work.

[miden]

sorry Microsoft Surface Pro (forgot the S)

[miden]

One quarter-watt film resistor value of 1.2 kOhm. Eighth-watt would work as good but they get kinda tiny. Probably 5 percent tolerance resistor is good enough but you could use 2 or 1 percent if desired.

One capacitor value of 0.68 uF or 680 nF. Which is the same value. MicroFarad is 10 ^ -6 and NanoFarad is 10^-9. Less desirable capacitor types for this use include electrolytic, tantalum and ceramic. They would probably work but are less desirable. More desirable types would include polyester, mylar, mica. About anything that ain't electrolytic, tantalum or ceramic. If you find one in a junk box or cheap on a store shelf with real high voltage rating it will work fine but for this use any voltage rating bigger than 25 volts ought to be fine. 5 percent tolerance ought to be good. Probably not worth the money to hunt down a 1 percent cap. 10 percent tolerance would work but just be a little less "predictable".

Your cable has a center conductor and a shield wrap.
_1 You would wire the shield wrap to the plug SLEEVE as normal.
_2 You would solder the center conductor to one wire of the cap and then solder the other cap wire to the plug TIP.
_3 Solder one end of the resistor to the plug TIP and solder the other end of the resistor to the plug SLEEVE.

If you can find some kind of little vise or maybe even use some grip-lock pliers to hold the plug in a convenient attitude for soldering, you would probably first solder one end of the cap to the center conductor. Then you could poke both the other capacitor wire and one of the resistor wires into the plug TIP terminal and solder them both at the same time. Similarly poke both the shield and the other end of the resistor to the plug SLEEVE terminal so you can solder them both at the same operation. It is good to use little bits of tape or heat-shrink tubing around the center-to-cap-to-TIP terminal stretch to make it unlikely that those bare connections can short out to the SLEEVE terminal.

Hi JCJR,

One thing I forgot to ask is if the cable I use is stereo, do I simply double up the components and add a set (1 resistor, 1 cap) to the + and - (or Tip and sleeve cables)?

[JCJR]

Hi miden. Yes for stereo you would make two copies of the cap and resistor, with the "ground" wire of each resistor going to the sleeve ground.

If this is actually ring tip sleeve balanced rather than stereo, I think you could wire it the same as stereo above.

Dunno if you ever checked the input impedance of your amp. But the cases I spot checked for modern crown and qsc pro amps listed 10 kOhm for unbalanced and 20 kOhm for balanced. Which means each "leg" of balanced has 10 kOhm impedance, so the suggested RC values should oughta work either way.

Could also be wired dual like that for an xlr balanced connector.

For balanced maybe a differential filtering wiring could be made that doesn't need to reference the ground lug, but likely only relevant for xlr, and the only advantage for xlr might be less susceptibility to ground loops in rare edge cases. Or that's my guess anyway.

[miden]

Hi miden. Yes for stereo you would make two copies of the cap and resistor, with the "ground" wire of each resistor going to the sleeve ground.

If this is actually ring tip sleeve balanced rather than stereo, I think you could wire it the same as stereo above.

Dunno if you ever checked the input impedance of your amp. But the cases I spot checked for modern crown and qsc pro amps listed 10 kOhm for unbalanced and 20 kOhm for balanced. Which means each "leg" of balanced has 10 kOhm impedance, so the suggested RC values should oughta work either way.

Could also be wired dual like that for an xlr balanced connector.

For balanced maybe a differential filtering wiring could be made that doesn't need to reference the ground lug, but likely only relevant for xlr, and the only advantage for xlr might be less susceptibility to ground loops in rare edge cases. Or that's my guess anyway.

Thanks! Cable is just standard XLR

[Jim Y]

I suggest using the high pass RC filter tool on this site...
http://sim.okawa-denshi.jp/en/Fkeisan.htm
The calculators allow multiplier symbols so you can enter "10k" for a resistor instead of 10,000. The capacitor sizes in audio filters are usually in NanoFarad values - or nF so you can try a value such as 47n for the capacitor (which is 0.047uF and you can also enter it as "0.047u"). Lower down the page it will produce the cut-off frequency of the filter and draw graphs.
In the XLR input, you already have R. It's the 10k input impedance. All you need is a capacitor in series with each signal input pin. 47n capacitor will give a roll-off starting at 338Hz . 47nF/0.047uF happens to be a common value used in guitar tone controls, so you can find them easily enough from guitar parts suppliers (although don't pay for the overpriced "mojo" capacitors some of them push).

[miden]

haha thanks, and I know you mean well - but WAY too much info there (as in at the link you provided there) for me hahaha!

[JCJR]

Jim Y has good ideas.

The "ideal" would be to build a box with a power supply, and input buffers with high-z input and low-z output which drive the filter network, and output buffers with high-z input and low-z output which drives the output jack. Also if throwing that many parts at it, one could also design a fancier filter "in between" the input and output buffers.

The main reason for the active input and output buffers would be to assure "near identical filter performance" over a wide range of source driving impedance or dest sink impedance. When the source impedance is "trivially small" compared to the filter impedance then the source impedance doesn't have much effect on the filter response curve. Similarly if the destination impedance is "much larger" compared to the filter impedance then the destination impedance doesn't have much effect on the filter response curve.

If the source or dest impedance is close to the filter impedance then we can't ignore the contribution any more, and must calculate a more complicated filter circuit including the influence of the source and dest impedance.

A common "rule of thumb" is that a source-vs-dest impedance ratio of 10X or better is enough so we can more-or-less safely ignore the driving circuit and the destination circuit. That is why I tried to figure an RC hipass at the geometric mean of impedance, "in the middle" between typical source and dest impedances. Which was about 8.2X or something.

Just using a capacitor and relying on the amp input impedance for the R part of the RC might work great maybe most of the time. The tuning of such a filter would depend on the speaker/amp input impedance. If it happens to be lower than 10 kOhm then the filter cutoff frequency would get detuned higher, and if the amp input impedance happens to be higher than 10 kOhm then the filter cutoff frequency would get detuned lower.

Using a ground-referenced RC "in between" the typical src and dest impedances, would hopefully "nail down" the filter performance over a wider range of src and dest impedances, but to REALLY nail it down you would want to build a real box with a power supply and active input buffer electronics and output buffer electronics. Trying to nail it down in the mid-impedance point is just a kludge attempting to get as close possible to a buffered circuit performance with only simple passive components.

If you know for sure the amp input impedance will ALWAYS be 10 kOhms (or whatever) then maybe just a series capacitor would work in all cases. Or maybe just a series capacitor would work in most cases. It has been too long and I do not understand the possibilities.

Except for knowing that the manufacturer might say the amp input impedance is 10 kOhms, the rest of the amp input might as well be a black box. Maybe it is a center-tapped 20k transformer? Maybe it is a 10k terminating resistor to ground, and a 100 MegOhm tube or fet or fet input opamp voltage follower? Maybe it is a 10k resistor into the "virtual ground" at the inverting input of an opamp? Maybe it is a 10k resistor into a very low-z bipolar transistor junction?

Perhaps in all of those cases, so long as it is 10k, then just by adding the series capacitor you could guarantee exactly the same filter curve? I don't know one way or the other whether all of the possible amp input "black boxes" terminated by 10 kOhm would respond the same way to a series capacitor. Maybe in every case it would respond the same, dunno.

[egbert]

This thread is a little odd - two posters - lfm and winstontaneous - have raised the rather germane issue of the monitor being on the floor. They get no response.

Bass frequencies are boosted if the monitor is near a boundary (wall, floor, ceiling). This effect is further doubled if the monitor is near two boundaries ( eg wall and floor) and it is doubled again if it is in a corner (ie near three boundaries - eg two walls and a floor).

In these three cases, the bass response which might be flat if the monitor is on a tall stand outdoors (ie in free space) is radiating into half space, quarter space and 1/8 of space in these three cases and the bass boost in these three cases is 3dB, 6 dB and 12 dB respectively.

So getting the mofo off the floor and away from corners could solve this issue.

[Jim Y]

I saw that "on the floor" too, but if it's a wedge monitor then that may be where the user wants it.
You can remove some bass extension simply by blocking the cabinet port, but with a powered monitor, they usually rely on the port airflow to pump some cooling air around the amplifier. If it has external heatsink fins, it probably won't mind a blocked port.

[egbert]

Many of these monitor wedge products - eg Yamaha DXR series etc - have presets for each location - on the floor as a wedge, standing and on a stand. They apply the required EQ to deal with the issue I have already covered.

[miden]

I did not really refer to the "floor" posts, as,well, it's a floor monitor! If I wanted a stand monitor that's what I would have bought. For the record I am totally aware of increased bass resonances when in close proximity to walls, corners, floors!

And if the monitor I have had "presets", then I'd not have created the OP, ne c'est pa?