From don@till.com Wed Dec 13 10:16:09 1995 13 Dec 95 13:16:03 +0500 4 Dec 95 03:26:26 +0500 Date: Mon, 4 Dec 1995 00:21:13 -0800 Subject: Re: Buffer questions (was Re: CEM3394/SixTrak Output Mod) From: Don Tillman Organization: Don's house, Palo Alto, California Date: Thu, 30 Nov 1995 13:03:25 -0800 (PST) From: "J.D. McEachin" (Since you asked...) I don't know much about the 3394 or the context its being used in, so take everything I say here with a grain of salt. On 30 Nov 95 at 11:18, M. Christopher Jones wrote: > Good question... I have found that the audio output from the 3394 is > quite loud (some have commented that it is undesirably hot) but it > hasn't become a problem for me, I simply adjust the gain on my > mixer. However, throwing in a 100k resister in the output as a > buffer would quell your fears and take care of the excessive volume > as well. A 100k resistor is not going to do anything to protect the chip. It could actually make things worse. No, a 100k resistor in series with the output will protect things to a few thousand volts or so. However 100k is an unacceptably high output impedance. For this situation a resistor divider with an output impedance of 1k to 5k ohms and protection diodes to the power supply should do the job just fine. If you're really paranoid, a single FET source follower stage would be nice. What you want is a low impedance path to ground. Therefore, you want to use an op-amp set up as a Voltage Follower, w/ a 1k resistor on the output. The output of the op-amp looks like a ground to anything misconnected to the output. +---100k--+ | | | |\ | 3394 -100k-+-|- \ | OUT | \__|__ 1k__OUT | / +--|+ / | |/ --- - Several problems here: The circuit you drew isn't a voltage follower, it's an inverting opamp stage; a gratuitous inversion isn't good for folks who depend on the polarity of the signal; an extra opamp in the circuit isn't a good thing; and finally and most importantly, I don't think it's a good idea to suggest that folks blindly add opamp circuits into a synth -- all sorts of things could go wrong (oscillation for instance) and without the tools and skills to detect it you run all sorts of risks of fried tweeters. (This is an age old problem with email conversations; you don't really know the skill level of the person you're giving advice to. My approach has been to word the proposal in a "high level" yet informal way so as to not tempt those without the appropriate skills to build it, but still provide meaningful information. I may or may not be successful, but that's the intent.) I'd set up the input similiarly, though I'd replace the 1k w/ a 100k. This time, the - input of the op-amp looks like a ground. This appraoch has an unacceptably high output impedance. -- Don