[sdiy] 1S2473 vs. 1N4148 diode specs
synthusiast at gmx.net
Sat Jun 9 23:15:12 CEST 2001
Ok, for a moment let's forget about the 0.7V drop theory and have a look at
some practical values: I got my DMM hooked up to my Polysix and I measured a
voltage drop of 0.4V across the diode (D19) that runs from the 5V power
supply rail to the RAM memory. And the most interesting part: the 1N4148
diode, in between the Lithium battery and the RAM power line, gives a 0.25V
voltage drop when the synth is turned off.
Yes, of course the circuit could've designed better from the start in
Japanese and other equipment, but in practice the voltage drop isn't as
large as one would expect from reading the specs of the 1N4148's and
probably of equivalents too. And yes, as you have proven, one could use a
Schottki diode for a battery backup circuit which works for most
temperatures the synth will get to endure and I guess it will even have a
smaller voltage drop than the 0.25V of a 1N4148. But hey, the circuit with
the 1N4148's is cheap, has worked for ages and one can play an open air
summer synth marathon concert in Vegas for a month without worrying about
reverse current leakage ;))
> I do not agree....
> In a LOT of japan designed battery back up circuits... they run
> the RAM one diode drop below the 5V rail. So the normal memory
> voltage is about 4.3V.
> That is a LOT of headroom loss...
> There are two ways to eliminate this...
> Raise the supply voltage to 5.7V... and supply the 5V circuit via a single
> diode drop... and then use a similar connection to the RAM... so the diode
> drops cancel.
> This is the way it SHOULD have been designed from the start... its only
> extra diode.
> Method #2... use a SHOTTKY DIODE such as 1N5817, with very low forward
> Do not worry about reverse leakage... it will NOT harm the battery. The
> reverse leakage is very low (I admit it is orders of magnitude higher than
> the 1N4148... but it is still too low to harm the battery)
> The reverse leakage only gets bad at high temperatures... which the synth
> will never see for any length of time... maybe at an all day concert, in
> sun, in the desert, near the equator, with high line voltage... you get
> idea. Even then its only 12 hours max...right.
> OK maybe a three day concert as above.... ;^)
> Also note that reverse current is only possible with the synth ON... mine
> off more often than on.... ;^(
> I have done this on all my synths and get less memory corruption, and am
> still running on the same batteries....
> The Shottky diode has a drop (in memory use... almost zero current) of
> H^) harry
> >From: "Synthusiast" <synthusiast at gmx.net>
> >Reply-To: "Synthusiast" <synthusiast at gmx.net>
> >To: "Synth DIY" <synth-diy at node12b53.a2000.nl>
> >Subject: Re: [sdiy] 1S2473 vs. 1N4148 diode specs
> >Date: Fri, 8 Jun 2001 22:13:35 +0200
> >Hi again all,
> >Thanks for all the responses. :)
> >I was reverse engineering the battery backup circuit of the Roland
> >and wanted to know if the 1S2473 diodes Roland used for this circuit
> >be replaced by commonly available 1N4148 diodes. And the answer to that
> >yes, though the path is a little long. I was told the 1S2473 is
> >with the ECG519 diode and I found out the ECG519 is compatible with the
> >1N914 diode, which is compatible with the 1N4148 diode ;) So the 1S2374
> >plain normal diode and couldn't even be a Schottki diode, as I learned
> >reverse current leakage of those diodes would ruin the non-chargeable
> >Lithium cell when used in a battery backup circuit.
> >Why I wanted a battery backup circuit in the first place? Well, I
> >Old Crow's battery replacement for my Polysix and was wondering if the
> >voltage drop could be reduced by using different diodes than the 1N4148 I
> >used. But as even Roland used these kind of diodes, I'll keep my 1N4148
> >the circuit and be happy with it :-)
> Get your FREE download of MSN Explorer at http://explorer.msn.com
More information about the Synth-diy