stevie

It could be lots of things: something loose, something leaking, something worn, something rattling. You can't really troubleshoot this kind of thing from a distance, but test tones can help pinpoint the problem. Download the 30, 40 and 50Hz tones from here - http://www.testsounds.com - and play them through the cab. You should quickly be able to locate where the problem is coming from.

https://www.google.co.uk/search?q=1U+class+bass+amplifier&ie=UTF-8&gws_rd=cr&ei=Ny6oWIaKLcbHgAbO9JHYBA#q=1U+class+d+bass+amplifier

Just make sure you use a proven design or get some advice before you cut the wood. This thread - and I think the OP will agree - is a good example of how not to DIY.

I'd hardly compare those Eminence drivers with a Ferrari.

To use a single 100mm port, make it about 70mm long. [EDIT] Just double checked this. Anything between 70 and 110mm would be OK.

Two 75mm ports, 100mm long will tune the box.

Here is the crossover circuit (it's not very clear, but the first inductor is .6mH and not 6mH): [url="https://postimg.org/image/nfy4sbfz1/"][/url] Film caps throughout. 1mm wire on LF coil, .7mm on HF coil. 17W resistors on HF. 30W minimum on LF (2 x 17W/15 ohm resistors would be OK). If you need a layout, I can do one, but it's quite a simple circuit. Here are also a few measurements I took of the finished system. They are anechoic measurements without any smoothing. The curves that look a bit rougher than others were taken further away and are not 100% anechoic, as they contain some reflections. Ignore the droop to the far left of some of the measurements – it’s an artefact. Showing the tweeter response. Crossover point is 3kHz. [url="https://postimg.org/image/yp4m6xw9b/"][/url] Impedance. This is a very easy amplifier load. [url="https://postimg.org/image/4c8vjwbyp/"][/url][url="https://postimage.org/"]imgurl[/url] Response 30 degrees off axis (top curve is on axis). You can see that the frequency response collapses between 2 and 3kHz because the bass driver is beaming too much here. [url="https://postimg.org/image/7ezc5xd6b/"][/url][url="https://postimage.org/"]uploadimage[/url] This is what you can expect if you are of average/above average height standing 1 metre in front of the cab (upper curve is directly in front of the cab). The response of any 12” driver starts to collapse at 1kHz off axis, and you can see that happening here. However, the output from the tweeter helps compensate for this drop from around 2kHz, which will help the player to hear what he is playing. [url="https://postimg.org/image/x963q3yk7/"][/url][url="https://postimage.org/"]upload pics[/url] I would stress that this is not your usual boom/tizz tweeter solution. Here, the tweeter really helps to define the sound of the bass. Once you have played a system like this, you are unlikely to go back to a single driver design. I’m trying to get drawings sorted now so that anyone can build one of these. I’ll post as soon as they are available - hoping for end of the week. I have also finished the design of a 12 + compression driver and horn system that Chienmort intends to build. It uses exactly the same cab and bass driver but has a Celestion compression driver and a CD horn, which will obviously cost a bit more than this tweetered version. The crossover is a lot more complex because of the CD horn and lower crossover point. If there is any interest, I'll post details of that here too. Finally, it would be really helpful if anyone who owns a top-end 12” bass cab and is not too far from Dorchester could arrange to pop in to see me for half an hour so that we can compare that cab with the Basschat 12 I have here, and also take some measurements – whether for publication or not. Just message me and we’ll get something sorted.

This is what the inside of the cabinet looks like now. [url="https://postimg.org/image/t91t3ld7t/"][/url] I’ve been using felt on a lot of my boxes recently (a wool/cotton mix) and I’ve found it works exceptionally well. I did try foam and BAF wadding in the cab, but they weren’t as effective. As usual, the material is inexpensive and readily available from eBay. £6.99 (delivered) gets you a square metre, which is enough for two cabs. I’ll list information on where to get all of the bits and pieces when we're finished. Use one layer all round except for near the port, with two layers on the bottom 25% of the back panel. This arrangement was arrived at by systematic measurement and is not just a best guess. The felt needs to be glued to the cabinet panels, with a few staples to keep the second layer in place. Contact adhesive is best (just apply to one side). I found this spray can of contact adhesive in Poundland, which makes the job easier. It contains enough for one cab: Not much more to say about this, except that if you are thinking of putting damping into your current cabinet, this is a good way of doing it. Now that we have a sorted cabinet, we can start designing the crossover. I’ve put a handle on the side and this is what the cab looks like with the drivers in: If you’d prefer a black port, it can easily be painted black with something like Holts bumper paint. The two raw drivers we'll be working with measure something like this: [url="https://postimg.org/image/pik463nxt/"][/url]

You've looked at the specs of the RCFs?

I couldn't have put it better, Phil.

Yes, one big port, Luke. The idea is to minimise friction and turbulence within the port; the less friction, the less chuffing and compression. It's not so critical with large cabs, or general purpose PA cabs for voice, but with small, high powered bass cabs like this one, optimising the porting makes a *huge* difference to the output produced by the port - and reduces distortion audibly. A single round port is simply the best solution for minimising friction. They're called flared ports, MoonBass. I have a selection of them here, including some very nice Precision Ports 4" ones. Unfortunately, availability is a bit of a problem in the UK, but they do work really well. I considered using the Precisions, but their external diameter is about 18cm and there wasn't enough room on the baffle. Also, a single flared 4" port wasn't going to be long enough to tune the cab to the target frequency. The performance of the 5" we are using is equivalent to a flared 4" port. The ideal would be a 5" flared port on the back of the cab, but nobody makes one. If somebody could use a 3D printer to make some flared port ends cheaply, that would make me very happy. It seems like an ideal job for a 3D printer.

This is the port that eventually worked best and which generates very little port noise even at fairly high levels. [url="https://postimg.org/image/km8htctt5/"][/url] It’s a 120mm Manrose pipe (125mm external), available from Screwfix (and lots of DIY and plumbing stores). You can cut two ports from it and it costs £2.69. I tried various sizes and types of commercial port, but this was definitely the one. In the process, I cut and filled so many holes, my baffle ended up looking look like this: [url="https://postimg.org/image/x8756ba9r/"][/url] This is the hole cutter I used, which produced a really snug fit in one go. [url="https://postimg.org/image/3wg83i137/"][/url] If you don’t have one (which is admittedly very likely), you will need to use some kind of hole cutter or router to cut the hole. Unfortunately, there is no commercially available port in this size with a lip to hide dodgy woodwork. I am now making a start on the internal damping. [url="https://postimg.org/image/6q42ragcd/"][/url]

A bit too much chuffing for my liking, John. Sorting the porting out has taken a while and lots of hole cutting and gluing with glass fibre and filler. However, I do believe I now have the solution, which I will post later today. David, PU glue is great when you don't have accurate joints and need to fill gaps, but these braces really do need to be pressed very tightly to the cabinet panels for optimal effect. If there is a slight gap because the brace (or shelf) hasn't been cut correctly or fixed down tightly, it won't work properly. While the gap-filling properties of polyurethene will prevent vibration if the joint isn't perfect, you really don't want any gaps at all. The damping effect of attaching bracing comes more from stiffening the panels than any properties of the glue. In fact, you can hear the effect of the brace just by pressing it tightly on the panel without gluing - which is how I arrived at the positioning of the bracing for this box.

Yes, I would also expect plywood to be stiffer than the softwood I’m using, JPJ. The softwood is actually quite stiff on edge, but the big advantage for me is that I can pick it up from the local wood merchant and simply cut it to length. I’ve now carried out sine wave testing on the fully braced cab and this proved quite interesting. Without boring everyone with all the figures, I found that the spurious vibrations at the fundamental frequencies of the bass (40 to 100Hz) had virtually disappeared and that the level of vibration at all other frequencies was much lower than before. This will have a positive effect on the sound of the system. It will also ensure that amplifiers placed on top of the cab will not dance about or fall off – perish the thought! Now that the cabinet is so much quieter, the sine wave tests showed that some more work needs to be done on the porting. Test tones are very good at revealing problems that you’d miss just using a bass as a sound source. So the next job is to take a closer look at the porting arrangement to make sure it’s optimum.

I found that the best way to brace the end panels was across the longest dimension. When the glue dries the bracing will be finished. I needed two lengths of 2.4m batten to effectively brace this particular cab. I forgot to weigh the batten before I started cutting it up, but I calculate that it adds around 1.2kg to the overall weight of the cabinet. Using poplar ply strips or shelves might reduce this a little. I’ve carried out this exercise on plenty of cabinets before. It is a particularly good way of “fixing” an inexpensive shop-bought cabinet, as these are hardly ever braced adequately.

I haven't quite finished the bracing, but the cabinet walls are already impressively quiet. I've been using lengths of 12mm x 32mm softwood (as mentioned by JPH) , which are readily available from local wood merchants and the usual DIY stores. The advantage of using these is that they are cut perfectly square, which makes it easy to get a tight glued connection between the brace and the cabinet panels. The secret of getting the brace clamped tightly to the panel is to use wood screws and to pre-drill through the brace into the panel (not right through it, obviously). Use plenty of PVA and make sure it oozes out from both sides right along the length of the brace. The most important brace is a figure-8 brace just above the driver, as this cuts down vibration on the baffle and stiffens the cab close to its centre point. There is another semi-figure-8 below directly behind the driver with sufficient space to allow for the driver and magnet. With a birch ply box, that is probably all you would need. The remaining two braces are glued to the back and side walls in a "U" shape. One set positioned near the top of the cabinet, and one set near the bottom. The figure-8 configuration cross brace is not needed here because the top and bottom panels provide sufficient stiffness. I ended up with four circular or semi-circular braces fairly evenly spaced along the length of the cabinet. By the way, this bracing arrangement was arrived at systematically using the sine wave test tones from my Alan Parsons test CD. I still have to brace the bottom and top panels. Then I’ll report on the final results. [PLEASE NOTE: THIS WAS A PROTOTYPE. THE FINAL BRACING DESIGN APPEAR LATER IN THE THREAD.]

Feeling for vibrations with your fingertips works really well. Even with an accelerometer custom designed for this type of work, you have to set up and calibrate your measuring gear, glue the device to various areas of each cabinet panel, and save or print off your results. With your fingers, you can check the different areas of the cab in seconds. The only drawback is that it is subjective, but once you've done it a few times, it's very effective. I'm now sorting out the bracing for the box, which will take a few days. Has anyone tried the test tone MP3s yet?

Before we can even think about working with the drivers, we need to take a closer look at the cabinet to make sure it’s fit for purpose. So let’s install and connect the bass unit. I’m going to carry out some tests on the box, which you can easily replicate on your own cabinet at home if you wish. The standard material for wooden cabinets in the world of sound reinforcement is, and has always been, 18mm birch ply. Because they are stiff, cabinets made from birch need relatively little bracing (I say relatively....). The lighter stuff is different, however: here, bracing is mandatory. The lightweight spruce plywood that our test box is made of is unlikely to be as stiff or well damped as the poplar ply we will ultimately be using, and it is certainly nowhere near as stiff as birch. Bracing is therefore a necessary exercise and we should be able to transfer the lessons learned to the final version. The test CD I’m going to be using contains test tones at the usual frequencies. Playing them through any cabinet starting at 40Hz quickly reveals resonances and problems, if there are any, with the cabinet’s structure. You could say it’s a bit of a torture test. If you’d like to try your hand at this on your own cab, download some of these MP3s and give it a go: http://www.testsounds.com. It’s a sad fact that many expensive bass cabinets suffer from cabinet resonances, by which I don’t necessarily mean rattles. It’s often a commercial necessity to build cheaply in order to remain competitive. After all, who checks a bass cabinet for resonances before buying, or even looks inside it to see if it has been properly braced and damped? The problem is that a vibrating cabinet will colour your sound and detract from its perceived sound quality. Panel vibrations simultaneously add to and subtract from the sound of your bass, continually altering and masking its tonal colour – so it’s important to try to keep them under control. To make matters worse, the bigger the cabinet you have, the more difficult resonances are to control. Fortunately, the DIY builder is not constrained too much by the extra cost of building a well-behaved cabinet. Test results Not great. Not unexpectedly, the cabinet as it stands rings like a bell. The worst vibrations were on the baffle – so that will be the first area to deal with. This test is normally carried out using an accelerometer, which is a type of vibration detector. Since I don’t have one, I simply felt the vibrations on each of the panels with my fingertips, rated them subjectively for level, and entered them on a sheet of paper in tabular form. Rather than reproduce the entire table, which is a bit difficult on here, I can summarise the results as follows: Baffle Resonances starting at 40Hz and going through to 400Hz. The worst frequencies were 63 and 80, but 50 and 100 were also excessive. Sides Resonances starting at 100Hz through to 250Hz. Then a single resonance at 630Hz. Top Resonances at 80 and 100Hz. Then at frequencies between 250 and 400. Back panel Resonances from 40 through to 315 Hz, with particularly bad resonances at 80 and 200Hz.

That can't be right. My maths is rubbish! [Edited for a grocer's apostrophe]