This was a quick “let’s see what happens” kind of mod for my TTSH.
In their stock config, the TTSH loudspeakers are supposed to be fastened to the circuit board, which is then fastened to the front plate. Handy approach eg. for having the electronics as a entirely separate, single ‘module’ for servicing etc. But problem with this is that it leaves a ~12mm gap between the driver and the front panel.
From what little I know about loudspeaker design from before is, the driver needs some sort of a baffle for more optimal performance. What the baffle does is, it stops the air from circulating immediately around the loudspeaker edge as the driver pushes air outwards or inwards. So having the driver installed as a almost free-hanging part (as in stock TTSH) is the worst possible solution ever in terms of driver performance. So perhaps it’s no wonder a lot of TTSH builders complain about the loudspeakers sounding crap!
To illustrate, here’s a few loudspeaker baffle designs:
Just to see what amount of performance could be tweaked out from the loudspeakers with a bit of tinkering, I figured I’ll try modding the driver installation towards something like the type A, the flat open baffle. Trivial to implement really: Bridge the 12mm gap with some kind of a custom spacer + gasket and that’s that.
I had some leftover 5mm thick sheets of clear acrylic, so I figured I’ll stack two of these and add some 3mm thick EPDM gasket tape to the sheet which goes against front panel. The driver has this sort of paper-ish spacer ring on it, thus with the installation screws tightened the combination makes for a pretty ok seal.
Visually this isn’t the prettiest of mods: Depending on the viewing angle, you can see the shiny clear acrylic reflecting light.. Not forgetting the cut edges (white / light gray-ish). Perhaps black acrylic would’ve been better, but didn’t want to go buy some just for this purpose.
But what about the results then? A decent surprise I’d say :). Here are plots for comparison:
Most notably the presence area between 2kHz – 5kHz turns into a slightly flatter downward slope with about between -3dB and -5dB cut and the low range between 100Hz to 300Hz gets a nice boost of around +3dB.. Values slightly depending on at what Hz you compare the two of course. So esp. on the presence range, the sound is a little less harsh on the ears. Great results for half an hour spent drilling & milling, nnnice!
How I recorded the samples was by playing a burst of white noise from TTSH noise oscillator all volume sliders cranked up, then recording that using the X-Y stereo pair of a Roland R26 placed at a 10cm distance off the front panel and about lining up with the driver center / axis. In case you want to listen to the recordings yourself, get them here: stock and modded. The spot I recorded in was far from anything acoustically treated, so you can hear eg. the room reverb on the recordings.
Of course this mod doesn’t make for a very large baffle area improvement: there are eg. the volume sliders (and some other holes) near the loudspeakers, from where the air can still escape. But at least the route isn’t immediately next to the driver frame!
To further tweak the performance, Type C design would’ve been even more optimal. But I didn’t feel like putting more work to the build at this point, despite the promising results of type A design. AFAIK The type C also needs to be air-sealed and making a tight seal over the PCB would make the electronics less serviceable.. Yeah, let’s not go there.
Alright on to some finished surfaces with the TTSH build.. This entry is all about tolex, from someone with zero previous experience on applying / handling technique. If you ended up here looking for some kind of advice, this may (or may not) be a good read. You’ve been warned! :)
Do also check out what has happened in this project previously, links are at the bottom.
Despite all the cutting board & dust cover biz in december (and against all expectations) I also had a few moments to start sketching CAD plans for the road case of my Two-Thousand-Six-Hundred. These initial sketches evolved into a marginally final plan by early january.
By “marginally final” I mean that the design is perhaps “80% there”. I really see no point in putting more work to draw the design all the way, as some of the fine details are better off handled on the fly. Kind of goes with practicing wood work etc. :).. Here’s that design:Read More…