czwartek, 29 sierpnia 2013

Loudspeaker - part 3

Okay, so today I have found some time to work with the enclosure. It was covered with a thick layer of acrylic filler, that was applied some years ago. This filler turned out to be too brittle for my use, as I wanted to have all the angles as sharps as possible. In order to obtain that I decided to use a totally different filler, a polyurethane filler, that is two part filler. I chose Finish (Novol) to obtain as fine grain as possible, yet have a hard filler layer.

Firstly I sanded really hard the acrylic filler to remove as much as possible. To do that I sanded it hard with 60 grit paper until most of the filler was removed. Additionally I sanded some of the MDF front to fit the shape of the enclosure.

Finally after sanding and blowing all the dust out I sprayed the sanded enclosure with acrylic primer (Motip, cheap one). The primer serves two roles here - firstly it helps the polyurethane filler to stick to the surface, secondly the primer fills small cracks and holes on the surface. The second role of the primer will be very useful later, as I will apply the same primer on top of polyurethane filler, before I paint the whole damn thing.

Primed enclosure (on the left), and box before priming (on the right).

Using a water based primer had a bad outcome on the OSB. Although the OSB I use is said to be waterproofed (with same kind of wax on the surface) one piece of the boxed absorbed the primer and swelled - fortunately only a little. Unfortunately this was the top part of the enclosure, so anything here would be very visible. To make things right I sanded this part (again), even harder to sand the wood of the OSB, to hide the swelling and primed it again. Still, it is very rough but this will be hidden underneath the filler.

At this point (and all further points, regarding the paint job) I would like to thank Saskia from, that helped me a lot in understanding how things should be made. I bow before your master knowledge.

niedziela, 25 sierpnia 2013

Loudspeaker - part 2 - the project

As I described earlier the columns are designed as 2-way system in a closed box. Now to give more details:

Used drivers are:

Tonsil GDN 20/40 8 ohm

This is an relatively old design, but I got this speakers quite cheap, and decided to use them. The thing that I like in them is that they work pretty well in the middle of the audible spectrum, around 2..3 kHz. This is why I have implemented them in this 2-way system. 

Visaton SC10N

This a dome tweeter from Visaton, also not very fresh design, although seems to be quite reliable. The mid-woofer is 40 W rated and the tweeter is 100 W rated, so it should withstand lower-than-normal cut at around 3000 kHz.


18 mm thick OSB loudspeaker box is 35 cm high and 24.6 cm deep and wide. After excluding the thickness of the walls (36 mm) this gives an approximate volume of 14 liters. The final volume will be even smaller, as in this calculation I didn't exclude the volume taken by the crossover elements, inner reinforcements, tweeter etc. Finally the volume should be around 13 liters. This produces a response curve like shown below:

This gives band (-3 dB) going from 70 Hz to 3500 Hz for the mid-woofer in theory. Reasonably speaking, it should go as low as 80 Hz with ease. You can see a bit of enhancement (up to +4.5 dB) in the kick-bass region, that is intended.


Passive crossover network is a classical CLC third-order network for the tweeter and nothing for the mid-woofer (simplicity again). The two capacitors are 4.7 uF and 15 uF, inductor is 0.56 mH giving the cut-off frequency around 2700 Hz.

Phase issues

The third order filter, used for the changes the phase by -270 degrees - -3/4 of full period. This is not possible to be compensated by plugging the tweeter in reverse polarization as with second-order cross-over, as it would still give phase mismatch by 1/4. At 2,7 kHz (cut frequency) the full period of the wave is around 13 cm - the approximate distance between the mid and high drivers. They reach in-phase work at around 3.8 kHz. Unfortunately in the 2.7 - 3.8 kHz the mid-woofers SPL is still around 90 db/W/m. This issue will be further investigeted one the loudspeakers are done.

Making of

The boxes were assembled some time ago (few years), so I do not have any pictures from the process. When I got back to the project I started from gluing the front MDF panels using universal spray adhesive (universal meaning usable for gluing paper, wood and textiles). This was a bit problematic as both sides - OSB and MDF - were just soaking in the glue, and close to nothing remained inside. At first I have covered both sides with glue, waited a few minutes and tried to glue them together with no good effects. So I waited till the first layer of glue dry up totally, applied another thin layer as a primer and then, after 10 minutes applied the final layer, waited 3 minutes and placed in on the OSB. After that I put them under pressure (25 liters of water ;) ) and let to dry overnight. It seems it worked.


The OSB boxes are partially covered with acrylic filler (I really regret using this one, as this is no hard enough to get the corners right). I will apply some other filler on top of that later.

To get the speakers centers together possibly close I needed to cut out a piece of the tweeter. It didn't turn out perfect but still acceptable (tape is to block the dome of the tweeter from the dust etc):

And finally the mid-woofer (still needs to be put together):

Next part - sanding/filling/priming the boxes!

Loudspeaker - part 1

As I have stated some time ago I'm biasing my resources towards audio systems for some time. Additionally I have several - quite old - projects that I have abandoned few years ago and want to finish them now. One of this projects is a 2-Way loudspeaker system based on Tonsil GDN 20/40 (that additionally needs some fixing) and Visaton SC10N.

I wanted to have this monitors as simple as possible, even if this means sacrificing some performance. In case of this speaker the sacrifice is in the low end of the spectrum - bass. The 20 cm woofer will  not have exquisite performance under - let's say - 100 Hz od 80 Hz. But on the other hand it wil provide a dynamic sound, what is more important for me.

The enclosure is 20 liters, closed (simplicity). GDN woofer is used without any equalization or cross-over (simplicity), SC10N is cut at around 3000 Hz with a 12 db/oct short pass CLC filter.

Enclosure is made from 18 mm thick OSB with an additional layer of 4 mm MDF on the front glued on. The MDF front layer is due to the fact that you can not mill in OSB and I wanted to hide the speakers somehow deeper in the front.

Actually I'm watinig for the glue on front panels to cure. After that - filing, sanding, priming and painting. I will post some more detailed information later on in the next part - together with some photos and calculations for the system.

sobota, 24 sierpnia 2013

XVI century coif

As I am completing an XVI century Landsknecht kit I've decided to finish some stuff that was 'on hold'. The first thing - a Landsknecht coif. Not really varying from a normal coif from that period (from late XV up to - let's say - middle of XVI) only with slashings to make it fit to the overal style. This kind of head-wear is in use from, at least, XII century to XVI or even later, I do not really know.

This is the pattern that I used:

Usually coifs like this were made from linen, as they were only worn under other types of head-wear in order to soak the sweat etc. In my case, and generally in late XV and on, I wanted it to be a simple stand-alone head-wear. Additionally I intend to use it as a part of padding of the helmet I'll be wearing later (proto-burgonet or similar).

My coif is made from two layers of wool in the same colors as my doublet, only in opposite order. The most external layer is from dark-blue wool, the inner layer from red. The coif also has lining inside, made from white linen. Apart from the linings inside seams, everything is hand-sewn with linen thread. I guess it came quite well, especially with my lack of skills ;)
Coif with one side of the slashing visible - you can see the inner red wool
Slashing on the other side

And some details:

Overall I'm glad from the outcome. This means that I have The top half of the Landsknecht kit ready and need only to make the trousers (or re-use the regular late XV century ones that I already have) and shoes. This would conclude the textilepart of the XVI kit... for now I guess.

piątek, 16 sierpnia 2013

Mysterious hardware, part 3.

So finally I have obtained an ISA-equipped PC that would be relatively new (a industrial grade P4) that allowed me to install Win XP and LabView software in order to debug the card, or at least try to do that. The motherboard was kindly donated by an user and blog-reader nicknamed music. Thank you!

Using a simple LabView software I tried to write and listen to/from the cards addresses (that I know, as I can configure the 7 MSB by jumpers). It is quite easy to do that, as LabView offers InPort and OutPort commands that allow me to directly talk to the bus. As I do not believe that it can be so easy I have hooked an oscilloscope to the cards address lines and address decoder output in order to check if it's working or not. And it is.

Unfortunately I did not get any response from the card. Two options are: the card (or cards CPLD at least) is dead or that the communication protocol is more complicated. I'm in favor of the latter. For now I have no further idea what to do with this piece of hardware apart from trying to read the CPLD content. As I do not have any cable for Lattice ICs and any proof that this approach will be valuable I just put the card back on shelve. At some point maybe I will come back - to further work with it or scrap it, as there are plenty of nice and rare ICs there.

Some new ideas

Long time, no hear. Again. But now I'm back from vacation with some spare time (let's say) for tinkering in the lab. I have revised my plans a bit as my focus changed. The idea now is to bias some of my work to audio related projects, especially Pro-audio and studio techniques. Recently I have acquired some stuff so I'll be trying to put it together. Of course it does not mean that the rest of things-to-do will be set totally aside.

So for now, several things to add to my list:

1. Speaker box - needs finishing (filling, priming and painting - I will elaborate soon) and fixing the speakers (also will elaborate in the next few notes).

2. Amplifier - a classic LM3875 based gainclone. Only needs an enclosure of some kind, as I have all the parts (I think so) ready.

3. Headphone amplifier - project work currently. An op-amp driving a complementary (MJE340/350) pair with DC-servo.

4. Other stuff, like an vintage BOSS reverb clone, or Coron DS-8 drum synth. Lots of stuff.

See you soon!