/tilted/

Just the keyboard should be do-able. Let me know what style you want. I think I have a bit of a selection...

Sign me up! (also, if you have any spare encoders, perhaps I can haul them away for you...)

I have several C64s in (probably) working condition, in Au. They're in storage. I'd be happy to let them go if someone can use them. EDIT: As in, for only the cost of shipping...

I think that a poorly designed crossover is probably what went wrong in Doc Brown's Lab. (ca 3:22 if anyone has still managed to not see this film...)

There's a place down the road from mine that sells all sorts of old restoration stuff including recycled timber, wooden doors and windows, brass fittings... heaps o stuff. They're called "Steptoe's", and they're so close to my place that they better keep the damn noise down!! I just can't help but think of these two: http://www.youtube.com/watch?v=8yGSGuhAapI anyway, i finally managed to drop in today, and they've got everything I need for my console. So awesome.

You could always use 2 40x2 displays... :P

I usually use the other one. I connect one pin to the DIN board, then i connect the other one to GND.

I think I understand what you're saying now. The 100nf/330nF caps are not there so much to make up for a shortcoming of the electrolytic cap, but rather to filter a different area of the spectrum. I think off hand (feel free anyone to correct me) that having a tantalum cap doesn't mean you won't need the 100/330nF caps anymore. What the caps are doing here is providing a low pass filter for your PSU. Adding more caps of different sizes is akin to changing the slope of your filter.

OK, well for the dual-rectifier design, there's no reason you can't use your current transformer. It is essentially a multi-tap transformer with two taps, providing symmetrical voltages. The other good news is that you can use the board design you have currently, sort of. At the moment you have a CT transformer with the outer taps connected to the AC (^v) points of a rectifier, the (-) becomes your 0 volt, the (+) feeds some caps, a regulator, more caps, etc. (Pardon the appalling ASCII art)... "All you need to do" is connect another bridge rectifier's AC points to: (1) the CT of your transformer, and (2) one (NOT BOTH) of the other taps. then: the (-) of this rectifier is tied to the (-) of the other rectifier (as these are both for your 0 volt), and the (+) of this rectifier is fed to the filter caps, 5v regulator etc of your board. This would have a similar effect to simply adding a single diode between the CT and the regulator, but instead of being half-wave, it is full-wave (which is much more efficient, and will allow for more current, with less loading on your filter caps). This is of course only one of many ways to do it, but it's the way I've done it, and tend to do it.

aluminium electrolytic capacitors are better for DC filtering, because the basic usecase here is for removing the AC component, following rectification. Their HF performance is limited largely by the fundamental design, which has relatively high internal resistance / inductance. They make up for this shortcoming by being relatively cheap, and being available to higher voltage ratings. They are also a liiiittle bit more robust when these ratings are exceeded, compared to tantalum. (Tantalum caps have something of a reputation for exploding...) In a linear DC supply, the electrolytics are usually used in concert with other caps of lower value, down to around 100-350nF. These caps are there to filter out the HF noise, leaving the electrolytics as your main storage caps. If you want to use a tantalum in a power supply, there's no great reason why you shouldn't. (I'm assuming the 47uF is not the only filter cap in there...)

Well, technically. yes. It could be used to create a +5v and +9v supply rail. It would be a really weird way to do it though, and you'd be leaving yourself open to really cocking things up if you made a slip. Given that in said configuration, you'd also be running the "GND" through a regulator, I don't think it's really advisable. Personally, I've never had any problems with the old 'multitap transformer, several rectifiers and several regulators' approach. Maybe there was some compelling reason why this approach wasn't used here. Guess I'll have to re-read the thread.

Teeth? They're for wire stripping!

Yeah, sorry about the 6.3V / 6V3 thing. It's an old convention which probably doesn't need to be followed any more. I think the original idea was that if you (for example) make repeated photocopies of something saying 6.3V, or write it with a bad pen, the decimal point might get hard to read, and the figure becomes 63V, and errors creep in. If you write 6V3, you'll lose the letters before you lose the decimal place. This has been taken for all sorts of measurements, eg 1R5 = 1.5 ohms, 1k8R = 1800 ohms (1.8k ohms) Really not an issue in an online forum though... Anyway. I wonder if the problem here might just be one of pure circuit topology. By this I mean that the rectification and primary DC smoothing seems similar to that used in a bipolar supply, but the regulation is taken more from a dual rail configuration. Mostly these differences seem cosmetic, but if you think about it for a second... in a bipolar supply, maybe the 0V rail would always wobble around like this, and have this amount of AC on it, but since the regulators are referenced to the 0V rail, it isn't an issue. This brings us back to the flapping bird analogy. Perhaps the issue here is that the 5V rail is the one that is wobbling the most, and the 9V is not referenced to the 5V, but to the 0V rail. Perhaps the bird's two wingtips are nailed (referenced to each other), and the bird's body is moving. -So the 5v is swinging around in the middle somewhere. I wonder if the situation would be changed if it were a -5V - 0 - +4V supply, with a negative 5volt reg, and a positive 4volt reg?

Sclk and Rclk INTO a DOUT board should be fairly robust. By this I mean they are meant to take signals up to around 5V as part of their operation, so feeding them a constant 5v is not so bad. However, if you have shorted sclk and rclk to 5V, you could damage the PIC, by holding a driven pin to a voltage level. I'd imagine the PIC has a certain amount of protection against this... Anyway, if you had damaged the PIC, then sclk and rclk would not work, and no dins or douts would work as a result. Try one DOUT board at a time (as you have). The next step is finding out which board is bad, (the one which makes it stop working), and figure out what is wrong with this board. EDIT: Formatting...

If the pots in question are indeed the type in the photo you posted, you could pass the wire through the rivet-post, adjacent to each pin. These are electrically connected to the pin, and are, uhm, hole-shaped. Can something be shaped like a hole? As these rivets are right on the carbon track, you don't want to cook them. Any realistic powered soldering iron should be fine, just don't use an acetylene torch to solder them. (actually an acetylene torch would probably vapourise the solder, anyway...)

Dammit!! I thought he was back for a second there!

I KNEW I wasn't alone!

It's alright, I knew that. In my hook-up, the taps are: 0 - 6V3 - 9V4 0V and 9V4 are connected to the bridge, with the 6V3 tap to the 7805. Think hard before you jump on me for this, remember, I'm not using a centre-tap tranformer. Another way to say the same thing would be that the "0 tap" is to the 7805, while the "6v3 tap" and the "3.1 tap" are to the bridge. This may explain the excess of AC in my proto, I'll switch to a 0-6V3-12V6 (ie 12V6 CT) when I get home from work tonight.

The waveform doesn't look like clean AC. It looks more to me like the AC is being shunted by the bridge, when it gets above or below certain points. These points are around +/-10mV from the baseline voltage (I had the oscope set to AC coupling, because I don't seem to have an offset, so this is the only way to look up close). And the transformer output is rectified. it's just sort of done weirdly. I have finally figured it out. It's kind of like a bipolar suppy, where the reference is 5V, the negative is 0V, and the positive is 9V. this is essentially the same as a +4V,0,-5V supply. Which is something that occurs to me. m00dawg, when your supply bogs down to ca 4.5V, does your 9V rail stay dead-on stable?

Okie dokie. I made a prototype of this: on breadboard, with the following differences: 1: The transformer is a type 2155 (15V, 1A, multi-tap, taps used: 0v, 6V3, 9V4) 2: Some small differences in Electro caps (6800uF is 3x 2200uF, 330nFs are a 470nF (7809) and a 560nF (7805) - all I had at hand) 3: none of the post-reg caps are mounted yet, so I can get a better look at the regs themselves. Some interesting results. With the entire 5v section removed, 9v supply works as a normal unipolar supply (no surprises here). nice smooth output, even with no additional smoothing caps. With the 5v section inserted, the 9v reg input is a little lumpier, and looks more like smoothed half-wave, than smoothed full-wave. The 5v regulator input looks like truncated AC, rather than smoothed regulated AC (ie, more like a square wave, rather than a sawtooth-ish wave). This might explain the AC you described on your 5V rail. The 5v regulator output has more noise than would normally be the case, with positive and negative going ripples from the baseline voltage. This noise is not huge (ca 5mV) -mind you, these tests were done with no load, but also with no additional smoothing caps. With the addition of a single diode before the 7805: The noise on the output of the 5v reg is still there, but the negative ripples are gone (as would be expected), but the positive ripples are still there, in a spacing consistant with half-wave ripples. Pictures to come, once I find a place to put them.

Hiya. ** The Answer ** The sockets are not polarised. The crystal is not polarised. That said: ** A little More ** ICs are polarised, and I always mount the socket in the same orientation as the chip should be placed, so that I can practically place chips in my sleep (handy for me, I'm very vague). Crystal Oscillators (as used in the MBFM, MBPOKEY, etc. -4 pin devices) are polarised. These are basically a crystal with a couple of buffer/inverters attached. the buffer inverters are powered by +5v, hence orientation on these is important. ** Potentially confusing for newbies section ** The crystal on the CORE board can be placed in either orientation, but I usually place non-polarised components in the easiest-to-read orientation (ie, I place them so that, as far as possible, they can all be read without flipping the board around) - usually I line up resistors etc so that the bands read left-to-right in the same orientation as the PCB's silkscreen . This is called component dress AFAIK, and I do it because I want to, not because I have to.

Love it! Very nice work as always Sasha. Do be careful (everyone) with the oven cleaner. - That 'slippery' sensation is the result of a process called "hydrolysis", and essentially, the oven cleaner is turning the oils in your skin (and by extension, your skin itself, since it is supported by the oils) into soap. Nasty.

Eeewwww. I think that bird analogy belongs in this thread. :) Perhaps then, the problem is that we are nailing the bird's wing to a rotating tyre? It feels to me that the stability of the GND in this design is dependant on how much current is being drawn through the bridge and the 9v reg. Personally, I'd have gone for the old two bridges, pretend the transformer is 0-6-12v (for example), tie GND to 0v tap approach. But that's just me.

Yes. Absolutely. All CCs are 7 bit. In fact, you can't send a data byte higher than 0x7F at all using MIDI, as it will be interprested as a new status byte. -Also, some systems could spaz if they get a new status byte before the data bytes they were expecting. However, the GM spec does allow for paired coarse and fine CCs. eg: [tt]0 Bank Select (coarse) | 32 Bank Select (fine) 1 Modulation Wheel (coarse) | 33 Modulation Wheel (fine) 2 Breath controller (coarse) | 34 Breath controller (fine) 4 Foot Pedal (coarse) | 36 Foot Pedal (fine) 5 Portamento Time (coarse) | 37 Portamento Time (fine) 6 Data Entry (coarse) | 38 Data Entry (fine) 7 Volume (coarse) | 39 Volume (fine) 8 Balance (coarse) | 40 Balance (fine) 10 Pan position (coarse) | 42 Pan position (fine) 11 Expression (coarse) | 43 Expression (fine) 12 Effect Control 1 (coarse) | 44 Effect Control 1 (fine) 13 Effect Control 2 (coarse) | 45 Effect Control 2 (fine) [/tt] This is of course, only relevant if following the original GM Spec, which is *largely* irrelevant these days, since midi is used less and less as a means of sharing compositions... (ie, why bother posting a midi file on a bulletin board, when you can just put an audio file up on myspace, etc for all to hear?... You could do this coarse/fine thing on any pair of CCs, so long as your recieving device accepts what it gets. Now, the 1024 steps thing sounds like a limitation of the ADC on your knob-box. Most PICs (including the PIC currently used in MIDIBox) have 10-bit ADCs. You can of course scale these values up in a MIDIBox, and you can get a wider range of values (up to 0-16383 if using 2 CCs for 14-bit mode, as you described), but don't expect to get any greater number of values. Unless you apply some sort of interpolation, you'll still have 1024 steps, it's just that these 1024 steps will be spaced over the range 0-16383. So... yeah. uhm, what was the question?

OK. I've had a really good look now. I never assumed this was a bi-polar supply, but I was (and am) struggling with two main points of this design. 1) GND reference is coming off a rectifier. Interesting. Not sure it will work. 2) 5V rail (pre regulator) seems to have no rectifier at all. These two points suggest to me that respectively (1) The GND will have a lot of AC on it, and not really be a GND. (actually, a GND can be pure AC for all I care, so long as everything else is stable in respect to it.) (2) The 7805 will drop out / heat up a LOT, and not really be able to stablise voltage, due to lack of rectification. Also, capacitors on this line will be ineffective, as they will be charged and then violently discharged in equal amounts each half-cycle. Please tell me I have it all wrong. I just cannot see how this design is supposed to work, because it just doesn't look right to me, and I'm really not sure what is being achieved by not using a more conventional design. That said, I'm keen to try it, and I will do so in the next day or so. (I also own a scope, so I'll post any results I get.)