mk59200@metso.tut.fi (Kolkka Markku Olavi) (06/26/90)
In article <3800@memqa.uucp> r91400@memqa.uucp (Michael C. Grant) writes: >In article <3364.26832a9d@ccvax.ucd.ie>, b_haughey@ccvax.ucd.ie (Brian J Haughey) writes: >> I'm looking for information on D/A converters that could directly supply >> an output to a speaker from low to high volume. (Is such a thing possible ?) > >For example, supplying high power output from a D/A converter would require >high powered semiconductor circuitry, and about the only high-power >semiconductors available today are for power amplifiers and special >solid-state power-distribution fault-detection circuitry. As for a >monolithic circuit like a D/A converter, you could try building one >from discrete components, but I wouldn't recommend it. Power MOSFETs are fairly common these days and I think you could make a high power D/A using them and the new 1-bit (Bitstream, MASH or whatever) conversion technologies. Driving the FETs in high speeds needs some thought, but otherwise it shouldn't be too difficult, if you can get a standard chip that converts the 16-bit data into PWM. >Secondly, volume control is also impractical in a D/A converter. There are several consumer systems around with digital volume control, it's really not that difficult (you do it _before_, not in the D/A). -- Markku Kolkka mk59200@tut.fi
bill@vrdxhq.verdix.com (William Spencer) (07/13/90)
in article <3800@memqa.uucp>, r91400@memqa.uucp (Michael C. Grant) says: > In article <3364.26832a9d@ccvax.ucd.ie>, b_haughey@ccvax.ucd.ie (Brian J Haughey) writes: >> I'm looking for information on D/A converters that could directly supply >> an output to a speaker from low to high volume. (Is such a thing possible ?) >> What are the problems associated with this, e.g. would the power supply and >> scaling factors of the D/A have to change as the volume changes ? > > Sounds to me like you want a D/A converter, a preamp with volume control, > and a power amplifier. It wouldn't make any sense to do it any other > way, for practical reasons. Actually he's got a great idea, although possibly not for the application in mind. Voltage amplification is a major problem from the audiophile viewpoint. Some designers even combine vacuum tube voltage amps with transistor current amps. 2 possibilities: 1. Use a resistor ladder voltage output DAC that can handle the maximum output voltage. Except for FET's used as switches, not amps, the signal path is so far passive. Resampling ("oversampling") filters must be used to allow simplistic output filtering. Passive filtering is used. A voltage follower buffer drives a passive volume control. A voltage follower drives the speaker. 2. Apply to switching type power amplifier. In this case a "one-bit" type of DAC principle is used The difference is the conversion is at full power. Filtering is passive between amp and speaker. Resampling is required. Switching amps are sometimes called a "class-D" amplifiers. "Class G" refers to the same thing with a peak power extension circuit; this is not required because the dynamic range is known. The circuit is much different than the analog version of switching amplification however because there is no comparison with the input. RFI could be a problem. (2) would be possible to build at home, (1) is not. While I find this interesting, my interest is damped by the scary thought of digitizing my phono signal :-) bill S.