Mottlefeeder

I get the impression that we are running on similar but slightly different lines of thought, so I'm not getting the answers I was hoping for. Having said that, I'm grateful for the explanations you have provided. They have filled in a lot of gaps for me. With my bass, the fundamental is 12 dB down, and since there will also be a 60 Hz 18 dB/octave HPF in the pre-amp, I agree that the fundamental is not an issue in my design. The speakers are rated at 80 watts each, and in a cab of the appropriate volume (11.4 l), are capable of taking 100 W at any frequency above 57 Hz. The excursion plots show that with the HPF in circuit, the speakers are below Xmax at all frequencies at 100 W. The proposed cab is tuned to 64 Hz, which seems to give the best compromise of size, smooth roll-of and bass extension, but a port capable of handling 100 W at 64 Hz makes the cab bigger than I want, so reducing the port power-handling capacity is a compromise I'm investigating. As I posted earlier, the spectral analysis of my lowest bass notes shows equal volumes at 60 Hz and 120 Hz, so I was hoping someone could comment on whether it was reasonable to assume that a low 100 W note could be regarded as for example 45 W at 60 Hz, 45 W at 120 Hz, and 10 W for everything else. That would give me the 'science' I need to reduce the size of the port and still be confident that it would do what I wanted. Am I oversimplifying a complex problem? Also, I was under the impression that if I reduced the port power handling capacity too much I would get chuffing, but you suggest that compression would be the more likely outcome. What does port compression do to the response of the cab? David

Unfortunately it's my nature to analyse things to death, and I still cut wood to the wrong size ... Taking your second paragraph first, I've taken on board that the harmonic mix will change as I move around the fretboard, but I chose the low C because it is the note most likely to overwhelm a port tuned to its harmonic. With regard to what the amp will deliver, I'm assuming that at some point I will turn up the amp and it will start to clip - 5 inch speakers are not as efficient as 10 inch speakers, so this is inevitable. Turning down slightly from that volume, the amp will be delivering its maximum voltage to the speaker, and the signal will initially have the lowest two harmonics predominating as each note decays. Using that scenario, I'm questioning whether I can derate the power handling of the port, but you appear to be saying that a port overload for the initial fraction of the note can be ignored. So, if the aim is to avoid chuffing, is chuffing a continuous overload of the port, or just overloading during the initial transient? David

Following on from my thread about whether WinISD was giving me bad advice, I'm now trying to work out whether I need a port sized for full power, or something less that full power. I downloaded a spectrum analyser and played a low C through it (lowest available fretted note). The analyser shows that the signal at ~30 Hz is about 12 dB below the signal at ~60 Hz, which is the same level as the signal at ~100-120Hz. For a port tuned to 60 Hz, my question is, if the voltage applied to the speaker is mainly two frequencies, and one transducer is reproducing those frequencies, the power must be split between those frequencies, so is it reasonable to design the port to take a maximum of half the amplifier power at 60 Hz, on the basis that the rest of the power will be at other frequencies? David

Assuming that the noise limiter was installed because of complaints by the neighbours, you aren't doing the pub any favours by playing louder than they want. David

If they are really old, then they will run at 5 volts. However the recent crop of fast chargers can run at higher voltages depending on what they are connected to. Check the charger labels - they should tell you. David

Just a quick update, and thanks to all of you who responded. I wasn't sure whether to go make a rectangular port or buy in a round one, so (amongst other things) I measured the internal diameter of the port pipe I would use, and it is 63 mm, not the 68 mm default in WinISD. This means that my port will be shorter than originally thought. Also, in calculating how much room I had in the box, I forgot that the port comes through the box wall, which means there is less of it in the box. The combination of those two means that I have room to use a straight port, and it finishes more than one diameter from the back wall, which seems to tick all the boxes. One of the things I have learned came from this thread - http://www.diymobileaudio.com/forum/technical-advanced-car-audio-discussion/156355-winisd-slot-ports.html - in summary, WinISD gives the option of modelling a port (a) finishing flush with a baffle, or (b) finishing hanging in free space, but a shelf port is neither of those. The 5th poster reported that a port finishing by being boxed in by the enclosure base and two sides, needs to be about considerably shorter than WinISD calculates it. He misremembered it as 40%, and later corrected that to 20%. So, if any air movement restriction in the vicinity of the port causes the port to act as it it was longer, this suggests that if you have to put a bend in your port, then you need to have at least a diameter's length of port after the bend to avoid the pipe before the bend affecting the air flow. Thankfully not something I need to worry about this time. David

And replied. David

And I'm back in the room. My basses are all 5-string, and vary from two-pup designs to a bridge Piezo acoustic bass, so I'm not planning to do any further analysis on output vs frequency. I'm in the right ball park so I'll stop there. So, firming up the design, with an 18dB/octave HPF set to 60 Hz, and limiting the input power to 20 W per speaker, and using a vent with a 2:1 cross section, I get an 80 mm x 40 mm vent 160 mm long giving me a maximum air flow of 15 m/s at 65 Hz. So far, so good. For an overall box depth of 305 mm, with an amp on the back, and a grill on the front, I have an internal depth of 220 mm for a port of 160 mm. Is the port end too close to the back wall? David

Thanks for that - I'll be away for the next few days, so I may not respond to further comments. David

Just to make sure I understand - the output from the pickup is 6dB down at 30-40 Hz because there isn't as much signal to pick up, so to model that into a port calculation, if I have say 10 watts at 60 Hz, putting me on the limit of port air-speeds, I should check that at 30 Hz, the port can cope with 4 watts without reaching that air-speed limit? Talking of compromises, and refining my design slightly, I was aiming for a similar sound and volume to my 8 ohm Eminence basslite S2010 in a small cab, taking about 10 watts from one channel of a stereo amp. This two speaker box will do it, with two 4 ohm drivers taking 20 watts each - i.e. half the size of box , but taking 4 times the power to reach the same volume. The next step up in volume would be to add a second S2010 and run one per channel, taking 20 watts in total. But the increase in output is only 3dB, and I am now carting a mini stack around. The alternative next step is to use my bigger battery-powered amp - 80 watts into 8 ohms / 150 watts into 4 ohms. At this point, the single S2010 will deliver 114 dB but the twin speaker box, with the drivers in series, will only deliver 108dB, so it is 6dB quieter. Beyond that, a pair of S2010s will deliver 116 dB, but I would also be carrying a bigger battery (27 kg!), so it would no longer be a portable rig. On that basis, there is no point in designing the twin speaker box to take more than 20 watts per speaker - if I want it louder, there are better ways to do it. David

If the port calculations are based on area times length, that implies that the volume of air is the important bit? Yes, a sharp 90 degree corner will cause turbulence, but the prefabricated pipes have radiused corners to reduce that. David

And I'm back in the room... My existing 10 inch speaker box has a thin slot port, but very low air speeds, which may be why I have never heard any problems due to turbulence. Moving away from a thin slot in this design, to something like 1:4 or 1:3, and big enough to bring the airflow down below 17 m/s, gives me a port that is longer than any of the box dimensions, so I'm now looking at a port turning a corner inside the cab. One option is a drainpipe with a 92.5 degree angle, and the other is a rectangular extractor fan duct with a 90 degree bend. Should I be aiming for a 'middle of port' length as per WinISD, or a port volume as per WinISD? And are there any rules of thumb for compensating for whatever problems are caused by bends in ports? Thanks David

Thanks for your observations and effort on my behalf. I'm on holiday at the moment and have no access to WinISD but I will be checking it out when I get back. David

Thanks for that - a commercial 3:14 ratio of port height and width is a useful benchmark. Stevie 's suggestions of 69mm diameter or 50:90 don't sit so well in the baffle size I had in mind, although that may change as the design matures. David Correction - since there are two ports, the 'combined port' ratio is 6:14, or slightly over 1:2.

My original plan was to use a slot port between the two speakers, so it would double as bracing for the cab sides, so I may have to rethink that. My existing rig uses Eminence Basslite S2010 speakers each in a 0.9 cu ft cab with ports tuned to 60Hz (based on an Eminence design). They used twin 3 inch ports, and I went for a 12 inch wide 1 inch high 7.5 inch long port instead. I haven't had any problems with my sound at the volume I play, so I was aiming to get a similar response curve with this design. I play a 5-string with a low B, and often use an HPF at about 50-60 Hz, but I tend to go for punch rather than rumble, so that may be why I haven't heard any problems. David

Thanks for that - I'll check it out with the new figures. Win ISD suggests 330cc's per speaker, so with an internal width of 150 mm, and the amp and battery strapped on the back it's looking like this - Briefcase speaker calculations.pdf

I had in mind to build a 'Joe Blogs Briefcase' using FaitalPro 5 inch drivers, but when I feed their parameters into WinISD I'm getting port lengths of over 300mm which does not look right. Can anyone be bothered to input the TS parameters for the FaitalPro 5FE120, for a cab with two 4 ohm drivers and a volume of 9900cc tuned to 70Hz with a 12dB/octave HPF with a 30Hz turnover. My clculations give - For 30 watts I need a slot port 150 mm*20 mm by 132 mm long giving me 17 m/s air flow. For 80 Watts I need 150 mm*35 mm by 260 mm long giving me over 16m/s air flow. For 150 Watts (my maximum amplifier power) I need a slot port 150 mm*45 mm by 345 mm long giving me just over 17m/s air flow. Do you agree with these figures? Alternatively, can anyone tell me how they split the power between the fundamental and the harmonics so that they don't have to design for full power at every frequency? Thanks David

Rod Elliott recommends a connection to ground via a 10 ohm resistor bypassed with a 0.1 mfd capacitor. Covers most situations and doesn't not need a switch David

I'm hoping for something modular so that I can fit the amp into the speaker casing, and power it from 12 Vdc when there is no supply available, or 230 Vac when there is. In the briefcase, PJ drops the mains down to 12 Vdc and feeds that, or the battery into a +/-X volt SMPS to power his class D amp. On 12 Vdc you have one voltage conversion, but on mains you have two in a row. It's inefficient having two conversions, but with mains power it is not that important. Looking at what is readily available, you can use a pure sine inverter with a 230 Vac rig, but that is one voltage conversion on mains, and two on battery, so it is inefficient and it is using up your battery power. As an example, my Ashdown MyBass Mk1 is Class D and with an inverter, it draws about an amp at 12 Vdc while idling, which is about a third of the capacity of an easy carry 7AHr battery. You can also use an automotive booster amp with an SMPS feeding a class AB amp, which has a lower quiescent current, but takes more on peaks. In summary, an automotive class D amp fed from a 12 Vdc SMPS would be the optimal solution. Unfortunately they don't do them in smaller power output units, and with a bigger unit you are back into the wasted power problem again. David