Grilled B3 Theremin

Posted: 12/15/2008 5:53:00 AM
Thierry

From: Colmar, France

Joined: 12/31/2007

A young photographer from Switzerland sent me his brand new B3 theremin in order to fix it. He had bought a 230V AC/24V DC wall wart, connected and... nothing. I guided him through a French theremin forum to connect two 9V block batteries in series instead, same result: nothing!

Polarity wasn't/isn't an issue since newer B3s have a diode in series with the internal voltage stabilizer in order to prevent such accidents.

What had happened? I first checked his wall wart: Although its nominal voltage should be 24V, I measured an off-load voltage of 37V! That was too much and that's why the B3's internal voltage stabilizer (uA78L12A) had been killed, there was no more output voltage.

Before replacing it I wanted to make sure that the other circuit components had not suffered, so I cut the uA78L12A's output and connected a 9V block battery instead. Now I had some triangle waves on my oscilloscope when connecting it directly to the pitch rod and the volume loop. So I can hope that the B3 will work with a new stabilizer circuit.

Finally I added a warning to this document (http://www.tfrenkel.com/theremin/b3_europe_mod.pdf) concerning the off-load voltage of some wall warts and the low power consumption of the B3.

I'll be in Germany for work tomorrow and I'll buy there what I need to fix this B3:
a) A uA78L12A voltage stabilizer for replacement
b) A wall wart with 230V AC/12V DC which has an off-load voltage of 16V
c) A slug which will allow easy connection/disconnection of the wall wart

Let's hope that the theremin community will have a new member in Switzerland soon!
Posted: 12/15/2008 4:37:29 PM
djpb_designs

From: Escondido, CA

Joined: 2/6/2008

The LM7824 is rated for 40V max input voltage, but there are deratings for temperature and current draw.

You might want to add a slightly bigger heatsink and be sure to replace the silpad (or mica insulator and silicon compound). High-line-voltage conditions might still push this regulator to its limits even with the correct wall wort.

Don
Posted: 12/15/2008 4:58:57 PM
Thierry

From: Colmar, France

Joined: 12/31/2007

djb, don't you think that this would be oversized? The input current of a B3 is between 10 and 15mA... and the Fairchild uA78LXX accepts only a max. input voltage of 35V, it seems that the low current versions are not so stable.
Posted: 12/15/2008 5:58:46 PM
FredM

From: Eastleigh, Hampshire, U.K. ................................... Fred Mundell. ................................... Electronics Engineer. (Primarily Analogue) .. CV Synths 1974-1980 .. Theremin developer 2007 to present .. soon to be Developing / Trading as WaveCrafter.com . ...................................

Joined: 12/7/2007

If the maximum current drawn from the regulator is 15mA, then I would put a 24V Transorb accross the input to the regulator, and a small current limiting resistor on the input..

supply(+) ---[A>DIODE|K]---[RESISTOR]---->Regulator (+) input.

supply(-) ----[A>TRANSORB|K]--->Regulator (+) input.

Doing the above gives maximum protection from any spikes or off-load conditions.. If resistor is 68 ohms, then there will be about 1V drop across it at 15mA, add 1V for diode drop, 2V for regulator drop-out, = 12+1+1+2 = 16V absolute minimum voltage required from Wort.. If voltage exceeds 24V, input to regulator will be limited to 24V by Transorb.. Make the resistor big enough to handle dissipation for any continuous input voltage >24V.

Far too much trust is (in my opinion) placed on the reliability of these small regulators - they fail quite regularly (particularly if some obscure brand is used) - I have seen many pieces of equipment destroyed due to this 5p component failing.. I always put transorbs on the supply inputs - yes, they cost more than the regulator, but they are worth every penny!

Edit..
Using a unidirectional transorb, you can get rid of the diode if you wish (save 0.7V drop) as the transorb + resistor give reverse voltage protection (sort of - the resistor will start to smoke quite quickly - so a 'fail-open' type resistor is advised!)
Posted: 12/15/2008 6:13:26 PM
Thierry

From: Colmar, France

Joined: 12/31/2007

Good idea Fred, but weren't you surprised too, to find such a high off-load voltage (37 instead of 24V)?

OK, I just had a further look on this very special wall wart: "Made in Italy" - this explains all ;-)
Posted: 12/15/2008 6:38:18 PM
djpb_designs

From: Escondido, CA

Joined: 2/6/2008

Sounds like that Italian company's transformer supplier was a bit sloppy with the turns count! Or maybe there was a short between primary turns.

The 78Lxx series has a lower max. input voltage than the 78xx series. Either way it is best to not have more than about 3-4V above the output voltage, so 30V would be the best high-line max. But you also have to deal with low-line conditions and there you want to have at least 2V above the output voltage for the thing to be regulating properly.

Transorbs are a great idea.

Don

Posted: 12/15/2008 6:40:03 PM
djpb_designs

From: Escondido, CA

Joined: 2/6/2008

Sorry ... forgot we were talking about a 7812! That would be about 20V max for high-line, 14V min for low-line.

Don
Posted: 12/15/2008 6:44:22 PM
djpb_designs

From: Escondido, CA

Joined: 2/6/2008

Also you might want to put a load on the wall wart by itself to see what output voltage you get, sine the unloaded output voltage is higher than the loaded value.

A 750 Ohm, Half-Watt resistor should do nicely.

If the loaded output voltage is within reason, it might have just been a bad regulator and nothing more,

Don
Posted: 12/15/2008 6:47:24 PM
djpb_designs

From: Escondido, CA

Joined: 2/6/2008

And one final note to this situation:

Mains power is not always properly regulated. If you don't want to dive into modifying your theremins, a power line conditioner might be the best bet. Furman makes a lot of them ... but they are not cheap.

Don
Posted: 12/16/2008 10:44:00 AM
FredM

From: Eastleigh, Hampshire, U.K. ................................... Fred Mundell. ................................... Electronics Engineer. (Primarily Analogue) .. CV Synths 1974-1980 .. Theremin developer 2007 to present .. soon to be Developing / Trading as WaveCrafter.com . ...................................

Joined: 12/7/2007

"but weren't you surprised too, to find such a high off-load voltage (37 instead of 24V)?"

Yes, it is a bit on the high side! - but when it comes to transformers (or anything containing an inductor!) I am no longer surprised by whatever I see..

To me, the idea of sticking a crude transformer / rectifier output straight into a valuable piece of equipment, is folly.. This is the most important real-world interface the equipment sees - what comes in, unless one has adequate input protection, has a direct route to almost everything... A regulator short-circuit failure can take everything out.

A transorb on the input does not eliminate the risk entirely.. If the regulator fails SC then there could still be 24V applied to the circuit - but the regulator is far less likely to fail if its input conditions are safe.. Also, if one did get a regulator failure, the input (pre-transorb) resistor may drop enough voltage and limit the current to prevent catastrophic failure.

I tend to go for overkill - need regulated 12V, use a 15V regulated wort, and have a fusible resistor dropping 0.5V at maximum current requirement, and a transorb across the 12V regulator input. (I also tend to overkill on ESD - small neon bulbs connected between antennas and ground being my favorite - why risk applying thousands of volts to your sensitive circuits when you can limit it to 90V with a 25p component which only adds a few pF loading)

You must be logged in to post a reply. Please log in or register for a new account.