[sdiy] Reducing maximum current through analog switches in ASM-2ADSR

[Barry Klein]

How about putting 2 switches in parallel?

----- Original Message ----- 
From: "Simon Brouwer" <simon.oo.o at xs4all.nl>
To: "Synth DIY" <synth-diy at dropmix.xs4all.nl>
Sent: Sunday, November 23, 2008 2:48 PM
Subject: Re: [sdiy] Reducing maximum current through analog switches in 
ASM-2ADSR


> Tom Wiltshire schreef:
>> Simon,
>>
>> On 23 Nov 2008, at 20:16, Simon Brouwer wrote:
>>> What do you guys think? The minimum attack time would become about 3 to 
>>> 4 ms. Is that something to be content with that or should I aim for a 
>>> shorter attack time?
>>
>> 3 to 4 mS is still a reasonably fast attack. For really snappy envelopes, 
>> you need to be closer to 1 mS - think Minimoog, Sequential ProOne, SH101. 
>> But then again, lots of famous polysynths from the early era of 
>> microprocessor-based envelopes couldn't manage times anywhere near that 
>> short (Oberheim Xpander only updated the envelopes every 10mS, and the 
>> Waldorf Wave is notoriously sluggish too).
>>
>> Do you really need to reduce the current so much? Is the chip actually in 
>> any realistic danger?
> I consider it bad practice if a design allows exceeding absolute maximum 
> ratings, because you simply have no way of knowing if you can get away 
> with it. Of course if your DIY synthesizer breaks down it's not a disaster 
> like when you have to recall thousands of products for repair or 
> modification if your decision to cut corners turned out not so well. But 
> still...
>> I'd have thought that although the peak current might exceed the absolute 
>> maximum rating, if this is done briefly (say for 1mS!) it might not cause 
>> any permanent damage. As the capacitor charges, the voltage drop across 
>> reduces, and the current drops correspondingly. Hence the period when 
>> this excessive current flows is actually very short.
> All I know is that the manufacturer specifies absolute maximum current of 
> 25 mA, and they don't specify a peak pulse current (which you can see in 
> datasheets of diodes, MOSFETs etc.). Who knows what the thermal time 
> constant is of a switch in the 4053. It might heat up in much less than 1 
> ms.
>> Have you worked out roughly what the *average* current through the switch 
>> would be during the attack curve?
> It would be somewhat less than half the maximum (and that would still be 
> too much).
>> Alternatively, is there another version of the 4053 that you could use 
>> that has a higher current rating without increasing the on resistance?
> Maybe the Fairchild one, its datasheet does not specify an absolute 
> maximum for the switch current at all...
> However, 15V times 0.167A amounts to 2.5W which exceeds the maximum power 
> dissipation of 700mW.
>
> http://pdf1.alldatasheet.com/datasheet-pdf/view/50867/FAIRCHILD/CD4053.html
>
> -- 
> Vriendelijke groet, Simon Brouwer.
> | http://nl.openoffice.org | http://www.opentaal.org |
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>