TheAncientOne

Ther are some notes by Limor's famous German engineer here: http://www.ladyada.net/make/x0xb0x/mods.html Some of the mods are worth looking at too.

Back in the day, Marshall did a reasonable bit of extra trade supplying empty cabinets and amps for videos. I know that one stage rig had at least 50% of the amps plugged in but not running, (I know because I'd filched the EL34's to fix the amps they were using). Any other sound engineers want to own up to putting a mic on a lead guitarists stack, with a DI box, as 'insurance', then never turning the mic up? - or in some cases letting them test it during the sound check? Or putting about 10 mics around a drum kit, then working with 4? I was asked by one band in the 70's to 'make the lights flash on this thing', a Moog Modular they had rented to make the stage rig look good - because it was feck all use for playing on stage. It was a very early unit and would only stay in tune for for about 5 minutes. It was a really good working patch, sadly no-one ever heard it. Names withheld to protect the guilty.

Interesting point. My HK supplier, sent me the pairs he was using in clones, I went with his recommendation. The parameter measured by the bin code is Idss, the maximum current that the FET will deliver. Q39 2SK20A (Y), is used as a constant current source in the tail of the Q40, Q41 darlington pair that buffer the envelope generating capacitor. as long as the current is within range, there can be no problems. Idss is for a "Y" is 1.2 to 3.0 mA, so there is a reasonable variation available anyway. The Second FET, the one in question, is a high impedence buffer on the VCO timing capacitor. The rating for an "O" is 0.5 to 1.4 mA. The rating for a "GR" is 2.6 to 6.5mA. Since there is a 10K resistor in the tail of the FET buffer, the maximum current that could flow is 1.2mA. You can't overdrive a FET in this instance, to me it seems the "GR" has a bit more headroom. Unless they deliberately run the FET a bit non-linear, there should be no audible difference I will compare notes once I get my Sasha x0x up and running, My LadyAda x0x is not quite right yet - I'm getting a more perfect sawtooth, (no flat topped bit), and I haven't yet tuned up the filter resonance - though I did select the highest gain transistors, from the LadyAda kit for the filter, (already done for you). I'll report back as I do. If you figure on experimenting, leave the leads long on the FET to make it easy to remove for testing. If you want to save rare parts for where they affect the sound, some of the ordinary transistors can be replaced as they are just pulse forming or doing 'support' jobs. 2sk30atm_en_datasheet_030325.pdf 2sk30atm_en_datasheet_030325.pdf

Do not drill - most motherboards are multi layer, and you might drill right though buried layers and either short or break them. You need a 40 watt iron, minimum, preferably a bigger one. If you haver a friend to help you can use two irons at once. Check what type of solder was in use on the board, if lead free you need to be working with lead free. Add a small pool of new solder around the pin, this will hold heat and allow better flow/contact. Move one pin a little, then the other. When pulling cut off pins, see if you can get some ceramic tweezers, they don't steal heat from the pin. Temperature is not the issue here, the actual amount of heat that you can apply is. One trick to un-block holes you cannot suck the solder out of with a desolder tool, is to find a thin propelling pencil lead and slip it through the hot solder. It can be broken out when the solder cools, Hope this helps a bit

I've hit this one quite a few times. The answer is usually a hidden layer ground plane. You simply can't get enough heat into the board to melt the solder, the thick copper for the motherboard's power gridding conducts it all away. The problem is not getting ahot enough iron, but of getting one with enough wattage. I've had 2 irons at once on some boards. Good news is that I've only ever had one not start working again after about 15 cap change jobs. Abit slot 1 boards were notorious: they'd fallen foul of the famous capacitor electrolyte scandal.http://www.dashdist.com/1u2u/company/capacitor.html Some fancy electrolytics actually have some kind of weird formed aluminium pins, which are pre-tinned. Getting a flux to stick to that bare metal, (and which isn't so corrosive it ruins your boards), is going to be a problem. These are usually only large 'snap fit' ones used in big power units though.

Just been comparing prices too. $1500 for SE x0x, against 1250 Euro for the Rozzbox. The Rozzbox seems like much better value. I'm with Sasha on those flatted knobs: way too 70's HiFi for me. There are quite a few case making extrusion systems around. In the UK there is a long established system for building consoles Dorlec. I've used it for industrial panels, but I keep thinking what a cool modular synth frame could be done with it. A while ago, I made some cases using some of the sections from BACO, actually designed for aluminium framed double glazing. Nowadays there are quite a few sections made for constructing illuminated signs, which might be good. Locally I know an anodising company which means I could get some very nice colour finishes done. Maddest thing recently was to re-create the Fuzz-Face case, by doing a lost wax casting into a plaster mould. We used 2-part silicone to make the initial mould. The metal was from some old patio chairs, and a simpler early try was with drinks cans. There is something very 'elemental' about pouring molten metal. A word of warning: it can really mess you up, so do all the safety stuff, and make sure the mould is very very dry, I still have a mark on my leg from when I helped pour bronze for a artist in the 70's. A small splash went right through my protective apron, through my jeans, and stuck to my leg. I had to yell and carry on pouring, because I was one of the two people holding a crucible with 5 Kg of liquid bronze still in it. Now, a cast bronze case....mmmmmm.

Yes. I''d go for the Deionised water option too. Just do the keypad itself to start with. Incidentally, if it's a reasonable make, you can often get a new pad for £30 - £40. Wash down carefully and dry with a hairdryer. Another couple of useful cleaning mixes are IPA and water, and IPA and water with a drop of detergent like washing up liquid. For defluxing some boards, a first run with straight IPA, followed by a water/alcohol/detergent, clean, with a final rinse with a little plain deionised water, will get a factory grade result. Using 'washproof' trimpots is a good idea idea if you can get them. If the liquid has got between the layers on a membrane type, tou may be out of luck, the contact layer contains a lot of silver and can be eaten way by electrolytic action, leaving nothing behind. If the hard disc failed at the same time, then the liquid may have penetrated further. If the mother board has been hit, don't bother with any attempted repairs: just sell the display, case, battery and charger on ebay. Check the HDD with an external adapter to see if tha actual disc has gone, or the motherboard. (edits to correct typos - was in post gig groggy mode when I posted)

I've got one of the kits Stryd mentions. Not too hard to build, good value, though care is needed in tuning, (goes for most Theremins). I can send you a copy of the Jaycar manual if you like, though the kit is the best deal. You can buy the EPE editions for download here: EPE back editions

In the UK, get Isopropyl Alcohol, from the Chemist. It's good for de-fluxing circuit boards too, (I use it for that), and delicate parts cleaning. Medically it's the stuff they clean your skin with prior to injections, and it's also used to get rid of sticking plaster residue. I think, (though I might be wrong), that French 90% is like our methylated spirit but without the dye and some other stuff. You can't drink it, but it is good for cleaning and de-greasing. Pro paint shops do sell other solvent based degreasers.

Thanks Julien - I'd never have thought of that type of paint, but of course it is perfectly designed for the job. "Unfat" had me puzzled for a moment, then the literal translation got to me. Ah! "De-Grease". one of the problems of dictionaries. (You should have seen some of my early attempts to buy electronic parts in Paris... At one place the guy behind the counter used to give me a pad and a pen when I walked in, then I could draw diagrams. First breakthrough was the magic words "Une Bidule comme ça") I've been advised to lightly wet sand the ABS case, (I want to get rid of the textured finish anyway), and to use a plasic primer - I notice Tamiya do one. From my model making days, I remember that matt white primer was best under silver paint. Wet sanding avoids heat, which can release chemicals from the ABS that mess up the paint, apparently. My spraying will never be as good as the finish you or Sasha got, so I think I'll ask a car sprayer near here to do the actual painting. Boxes look good!

Got mine this morning - definitely worth it, panels absolutely beautiful. Enough to make me start talking to a local custom painter about spraying the Pactec Silver to match. Thanks for all the hard work, you put into this, Sasha. Let's hope for a more stress free, very acid, New Year!

Hmm. There might be a clue in the name too Holmes.....

Happy birthday Thorsten!

Regarding Stand offs, I used them with my flat-topped LED's, and got a good result, I've also done them using a flat panel taped to the front, covering the LED holes, assembled the board to the panel, with the LED's fitted but unsoldered, tapped them into place and soldered them - worked well on a few panels I did. For round top LED's I drilled a plastic spacer, matching the panel to get just the tight amount of 'lift'. The I/O panel sits quite firmly supported by the jacks, as you say. There is not a lot of weight or stress on it. That being said, Sasha and I have chatted about doing a board set which doesn't use so many fancy parts, and will work with standard panel pots, (a very good idea if you were talking one on the road). I envisage an 'analog' board, which could be used by analog synth builders on it's own, as a voice card, with options for a cheaper build without the dual transistors, and some mod space; a logic card and a 'front panel' board for the switches, LED's, encoder and display. The display being used with an additional small processor perhaps, as a way of getting rid of the Grayhill switches. Lots more hand wiring, but lots more flexibility too. Hope it's soon blippin' and trippin'

There are some errors in the on line doumentation. R269 and R270 are 10K. see: x0x Sequencer Build Page Hope this helps You can memo for any more details - I've got a LadyAda kit up and running.

I did a few experiments in that line too. Including a stack of old BSR 8 track players, fitted with NAB jingle cartridges. I keep looking for a clearence batch of Walkmen, preferably the auto reverse kind. Also did an EPROM based player, using u-law compression on 8 bits, 1 EPROM and an R-2R (sort of) ladder per 'note'. 36 pieces of veroboard..... and I haven't a clue what happened to it. Changed the, (short), samples by changing the loop base address. had, (I think), a pair of start 'instruction' start bytes to determine single shot or loop length. I wish I'd been able to use PALs and GALs for that - might be worth doing again that way, though I'd have a load bus and use Flash RAM these days. Harest bit then was agetting a good high speed sample clock. Definitely no glide or note bending available. Perhaps a modern version using something like SwinSID technology might be worth doing, could do a scanned keyboard and anote allocator to get away with less generators if need be, but I like the 'one per note' purity, somehow. My conclusion from all of them was that the hardware is not as hard as building the sample set.

The Mellotron 2008 Sound LIbrary The comments had me laughing out loud: Examples "This is a special combined sound made up for Mike Dickson, comprising (approximately) 40% viola and 60% M300A strings. The overall sound is extremely melancholy and well suited to the miserable sort of music that Mellotronists the world over like listening to, usually because their prog-rock fixations prevent them from getting any Hot Girl Action. " "Adrian had all the windows in the Empire State Building removed and 12" speakers fitted. He then set up a microphone in LA and let rip." "It has often been said that a Mellotron is one of the few instruments where you can get away with playing only one note and having it sound impressive. This is never more easily demonstrated than by the sound of the string section - as the entire career of a nameless German Electro-Ambient band playing hours of E-minor appears to show." There's a lot more - and samples.

If you need 4N37's - I have spares. I can easily do another on or two 1% resistor kits, (I nearly sent one, thinking about it), just let me know and I'll drop then in the mail today - it's not as if it's airmail. I may have 1 74AC174 to spare too, though I'll have to check. Glad all arrived OK.

Posted so far: Sasha MTE Goblinz Kartoshka AsafNetzer Kokipsiho Julien Voirin B-Funk Berg: ready and waiting, (wants a later posting).

I'm packing and invoicing now. First three posted today. Will have the rest invoiced before weekend, (Had run out of some 1% resistors). A couple of people will find that I've 'balanced up' their order - like they ordered enough Panasonic pots for 2 units, but only 1 set of Alpha's. Since I had some low cost heatsinks for the power reg, I threw in the ones I hadn't used on a 'lucky dip' basis. I've put a detailed invoice/checklist in with the parts. Mike

Though I haven't used it myself yet, (needs XP and I run on Windows 2000), Siemens "Solid Edge 2D" is free, and I have heard some good thigs about it from one of the local colleges. It will import/translate a wide range of files too. www_plm_automation_siemens_com - Solid Edge 2D

I got mine from Rapid in the UK. www_rapidonline_com - MEC Caps The ones I ordered were the 16mm 'thin' ones. The square ones, (expensive!) are good for some jobs, because you can open them up and put lettering inside. Too much for a MIDIbox FM, but I did use one for the Zero crossing switch on my Marc Bareille 'Quantix 8' m_bareille_free_fr - Quantix 8, I've not done the panel yet, but i'm trying for that old style computer/instrument look. A whole panel of them in a grid would look amazing for the FM, but nearly £40 worth of caps is too much for me, sadly.

Now you've done it! I have a big batch of Illuminated MECs, courtesy of an Electronics Goldnmine clearence and the very kind help of Peake. Just bought enough clear caps to try this: I also wondered about using separate buttons for the instrument/operator buttons, for compatibility. I think some diodes might see the thing working with very little code reworking, seeing the way TK did it. Key rollover might be a problem, though no worse than of two row buttons were pressed anyway. Have I caught the madness? Only time will tell!

Wonderful - chilled me out nicely. Thanks.

That's the conclusion I came to - you've mixed up the wiper and fixed connections, so that as you turn the pot, the full voltage is applied to a lower and lower resistance, eventually you exceed the current rating of the track and it burns out like a fuse. If you have a test meter, check that you have the right variable voltage coming from the wiper before hitching up to the AIN. Are using slider pots? These are much easier to make mistakes with, as there are many different wiring conventions for them. Don't feel too back - just about everyone building an analogue synth has done this at least once! My favourite trick being to fit the pots one those 1mm plus systems, and mix up the crimped on connectors in the body. My usual one is the backwards pot, though occasionally I manage the full swap. By the way. No-one seems to fit any small damping caps from wiper to ground on these things, which would certainly help cut the noise effects down. I need to look at this, because I've acquired a large batch of beautiful cermet mini pots with built in knobs, nearly free, that are just crying out to be made into a MIDIbox64. Trouble is that they are all 100K, so I need to look at the noise/jitter problem first.