This gallery contains 5 photos.
On request from a reader; here are some more pics showing where to connect the wires.
This gallery contains 5 photos.
On request from a reader; here are some more pics showing where to connect the wires.
A couple of months ago I started building a new turntable to replace my Pro-Ject Debut 2.
During the projects I have had a lot of help from the nice people at VinylEngine.com, and I also have a thread there where all information about this project can be found. I think I’ll migrate the substantial information to this blog later on.
I think this thread might be of interest to anyone dealing with turntables. I am a newbie at this area, and maybe my “thinking-out-of-the-box”-approach might have resulted in something interesting. The intention has never been to make the best turntable ever, but to make a better one than the last TT I had, and also learning a lot from it.
After I completed my “passive preamp”, there would be no need for the old Rotel preamp I had been using, if it were not for the fact that I also needed a RIAA amp for my turntable.
The amplifier itself
The design of the amplifier itself was not altered very much. I bought the components from HifiCollective.co.uk and made my own PCB design. There are a lot of choices when it comes to components. One could spend a fortune, or almost nothing on capacitors and resistors. I used Nichicon FG and ES capacitors for the electrolytics, and Takman metal film resistors. To be honest, I don’t really have any idea how the performance would be affected if I changed the Takman resistors for cheap NXP ones. Probably not at all.
The opamps are OPA2134, as suggested in the project description. It would be interesting to do some blind tests between different opamps. Luckily I installed them in (good quality) sockets, so they can be easily changed.
The picture below shows how the PCB for the amp looks. The two channels are actually completely separated. Only the power traces runs between them.
The amplifier needs a dual stabilized voltage power supply. Conveniently I happened to have a couple of 2x15V toroid transformers laying around. I am not very experienced with audio design, but got the impression that the LM317/LM337 adjustable voltage regulators perform pretty good in audio applications. Some regulators can be very noisy, and I guess a RIAA amp is particularly sensitive to noise because of the very low signal levels handled.
I put in trimmers to be able to adjust the output voltage. This makes the design a bit more versatile, and maybe I can use it in more applications later on.
First test run
First I temporarily wired the PSU and the Amp board up to do a first test – just to see if things would work as expected. Although I have a strong belief in Placebo, I really found the audio quality much better compared to when I used the RIAA in my old Rotel preamp. There were some noise though, and that was not very surprising considering the wiring.
At work, I found a nice aluminium profile from a pneumatic cylinder in the junk bin and started to figure out how to use it in the chassis. I ended up using two pieces it for the side walls. The top and bottom are made from 1.5mm sheet metal, and the front and the back are made from aluminium bars that I milled and drilled holes for connectors etc in. Have to say that it turned out pretty good in the end.
I put some screening around the amplifier board, and made sure to ground all parts at the same point to avoind ground loops.
After putting it all together, I did a lot of work trying to reduce noise. When I turned the volume up a lot, with the cartridge lifted up from the record, there were some noise that I did not like. After some work I managed to reduce the noise a lot. First, I changed the signal cable from the turntable (a Project Debut II) to a better quality one. I also twisted the wires between the RCA terminals and the amplifier inputs and that really made a difference.
Relocating the signal cables in respect to the power cables (outside of the amp) also made a difference.
When it comes to sound quality, all I can say is that it was a great improvement from what I had before (Rotel RC-960BX). It makes a good companion with the rest of my setup, and listening to vinyl records is a much better experience now.
There are probably lots of things that could be changed/altered in order to improve the performance, but I think I will just keep it like this for now.
The schematics and PCB design are available if anyone should be interested. Please do also feel free to send me a message if you have any input regarding this project.
Originally posted on august 09, 2009
A new speed sensor for my Buell XB9R
The speedometer sensor for my Buell XB9R broke the last day. The new ones are insanely expensive in Sweden (>1600 SEK ~= $220). They can be bought from GB for about £60 and from U.S.A for ~$50, but still it’s expensive since the quality of these apparently is really crappy.
So, I thought maybe I could make a new one myself.
Anyway… to be completely sure that the sensor was the source of the problem, I put together a simple “simulator” that I plugged in to the sensor connector (Buell side). It generates a few frequency sweeps, testing the speedometer instrument. Simple, but very handy
As you can see, this is only the bare IC chip, and has to be installed in some kind of enclosure. It also needs some electronic components to adapt the output to the speedometer and provide some protection against voltage spikes and such. These additional components will cost about $10.
I have made a prototype enclosure in some plastic material I found at work, and plan to mount this chip + components in that. The cylindrical part of the enclosure can be seen in the center of the top image.
At the moment, I am waiting for the sensor chip, + other components. When they arrive, I will do a test run, to see if everything seems to be working as expected.
I plan to publish the results of these tests here, including schematics and drawings. Please come back in a week or two for more information!
Update Aug 12th
Received the hall effect sensor today. Fits perfectly into the enclosure (first pic). Have soldered a cable to it, to be able to do a test.
Would be nice to have an oscilloscope now, so I could verify the signal levels before connecting it to the bike. Ordered one a few weeks ago, but don’t expect to receive it in another few weeks.
At least, I have done some tests, and it seems to be working fine:
Update Aug 16th
Put the components together. Using a PCB felt a bit overkill, so i soldered and glued them into a nice little ball.
I have used a simple LC low pass filter, a diode in series with the power source, and a 24v voltage suppression diode in parallel. Hopefully, this will be enough to protect the hall sensor from blowing into pieces when the power fluctuates.
Also, I found out that there is already a pull-up resistor (to 5v) inside the instrument module, so there is no need for this at the sensor (open collector output). Nice. Otherwise I would have to get 5v from somewhere (zener diode/7805 …).
Next thing to do is a temporary fixture for the enclosure, so I can mount it to the bike and do some evaluation.
Update Aug 17th
Made a temporary fixture for the sensor today, and made a test run…
If I were a smoker I would have a cigar right now. Think I could afford a rather expensive one for the 1500 SEK (~$205) I have saved. Well… maybe I should use it for a while before I draw any final conclusions, but it looks promising.
The pics below shows the sensor as it looked when I tested it.
Next thing to do is milling an aluminum enclosure, filling the plastic tube (containing the hall sensor) with epoxy and mounting the enclosure to it and sealing everything to keep it nice and dry.
I’ll be back…
Update Aug 18th
Made a groove for the o-ring, and filled the plastic tube with epoxy. To be sure that the entire tube was filled (the epoxy doesn’t poor very well) I used an injector tool. I also filled the tube i three stages.
Have also done some figuring about the design of the aluminum enclosure, and think I have a clear picture of how how it should look. Think I have to wait to the weekend before I can use the CNC mill at work. Anyway, I have to do the design in Inventor first and that is not really my primary element…
Update Aug 19th
Have designed the aluminium enclosure. Have to check that everything fits before milling it, but it should be no problem.
Took a day off. Will do the milling tomorrow
Update Aug 21st
Made the enclosure today and have to admit that I am pretty satisfied (pictures below).
Apart from a broken tap (when I made the fixture) everything went on smoothly. Will do the rest tomorrow.
Assembled the sensor today. Also made a cover to put on the top of the enclosure.
Once I have tried it out for a while, I’ll put some effort in the finish. Seems to be working well. Hope it will last a bit longer than the original sensor, but if it don’t – I can repair it for a few bucks.
Update Okt 12th, 2011
The sensor worked perfectly until I managed to crash the bike into a crash barrier. My new XB12s still has the original sensor, but as long as it works, I’ll keep it.
Originally posted on May 15, 2011
To be able to control the volume on my passive preamp via remote, I put together a receiver/decoder to be able to use my TV remote.
Started by finding out how to decode the IR data sent from the remote control. There are several different standards, and I chose the NEC standard since it seemed pretty easy to implement in a microcontroller. I bought a small credit card shaped remote from EBay, that I’m sure will work out fine.
Later, it turned out that the remote used with by TV also “speaks” NEC, so I can actually use both of them (there are a couple of free buttons on the latter).
I set up a timer in the microcontroller to increment a counter with a period of 280μs. This gives two increments per “base period” (the shortest pulse is 560μs). An IR receiver module, with some integrated functions sends out a TTL level signal, 0 where there is a 38KHz IR burst coming, or else 1.
The output of the IR receiver is wired to one of the inputs on portb of the PIC16F628A. Whenever the signal changes its state, an “interrupt on change” is trigged.
The time for high and low state is measured, and the bits are decoded. When 32 bits are received, they are split into address and command bytes and can then be evaluated in order to (for example) turn the volume up/down.
To drive the motor of the volume potentiometer, I use an integrated H bridge (LMD18200). This might be considered a bit overkill, but I had already ordered a couple of those for another project. Also, if I find out that the pot moves to fast, I can use PWM to slow it down a bit.
Yesterday I did a quick PCB design, manufactured the PCB and mounted everything but the LMD18200 (since I have not received it yet). Seems to be working fine. Will put up some pics later when the H Bridge is in place.
The design is in DipTrace format. Send me a message if you are interested. The source code for the PIC will also be available later.
The board has a serial port that can be used for sending out received keycodes. Also, there is a couple of pins on the board, directly wired to the PIC, that can be used to control other things than just the pot.
Kind of depends of what’s the PIC is programmed to do. Both the board and the source code is pretty customized for this application, but can be easily changed for other purposes.
Okt 12th, 2011
The board is installed in the passive preamp I put together earlier, and has been working well for a couple of months now. I lowered the speed of the motor a bit by sending a square wave to the LMD18200 rather than just a logic high.
Originally posted on May 08, 2011
Although not entirely correct, the term “passive preamplifier” is often used where a potentiometer is put between a signal source and the power amplifier. In other words; it’s not really a preamplifier at all, and in most cases (pre)amplification is not needed since the out level of most signal sources (CD players etc.) is high enough.
Since I realized that there is something wrong with my Rotel preamp (volume pot behaves strange), and I also really want to be able to adjust the volume from the sofa, I bought an Apls motorized pot från EBay.
The encolusure is a simple Hammond aluminium box, and the internal wiring is Van Den Hul Videolink 75. Not sure about the brand of the RCA terminals.
The pot is PCB mounted, so I made a simple PCB with screw terminals for the motor, and solder pads for the wires.
The knob is turned out of a piece of Impax (what I found in the junk bin). The painting is what I happened to have for the moment, and it turned out pretty good.
A remote control receiver and motor drive will be put together later.
Originally posted on mars 11, 2010
Have actually finished this project at this moment, but thought I could show a couple of pictures.
Started out by molesting an old AGFA scanner. Removed the old circuit boards and the mechanic stuff, along with some of the internal chassis parts in order to be able to fit everything.
Welded a box out of some sheet metal, painted it white, and then mounted it inside the box. Then I installed the UV tubes.
Of course, the UV box should also be equipped with a timer. A friend of mine gave me a couple of really cute old displays (HP 5082-7340). A PIC16F886 does the timing, and I used a 24PPR encoder with push-button function for the input.
Actually, I used the (semi-finished) box to make the PCB:s for it:
The result turned out pretty well.
The schematic and source code is available here if anyone should be interested.
Originally posted on november 21, 2009
My latest micro project; adding a coaxial socket to this vintage video game (Radofin Tele-sports), enabling me to connect it to equipment without a TV receiver.
This is how it looks. Beautiful, ain’t it?
Inside is kind of empty…
Did not take many minutes to find a coaxial signal.
Soldered a cable to the PCB and put a socket (gold plated ofc) on the back.
Yay! It works!
Me and my brother in some serious action
Even better on a projector!