I have this great tranny from a dyna ST120 that I want to use for a gainclone amp. The problem is that the tranny has only 2 wires on the secondary : 70VAC (no center tap). I know that running this through a full wave bridge rectifier will give me about +90VDC. Can I split that +45,0,-45 by using something like a 1k resistor from each side to create a ground between them? (I have seen this done with preamps run off a wall wart, but we are talking about a lot more amperage here). What sort of watt rating would the resistors need to be?
Thanks for the help!
Thanks for the help!
If you aren't drawing much current from the CT, then yes.
Typically, an amp only needs to draw AC through the CT (for the speaker ground return). In this case, a virtual ground can be made with a voltage divider, bypassed with large capacitors (a cap to +V and a cap to -V, never a single cap to one rail!). Remember to AC-couple the input too, otherwise you can make the amp apparently saturate, as the capacitors run out of voltage at LF.
Other possible cheap (or not so cheap) hacks include adding an autoformer to create your own CT, or using a large series inductor, to get the winding's DC voltage, filtering the ~35V ripple on it.
Tim
Typically, an amp only needs to draw AC through the CT (for the speaker ground return). In this case, a virtual ground can be made with a voltage divider, bypassed with large capacitors (a cap to +V and a cap to -V, never a single cap to one rail!). Remember to AC-couple the input too, otherwise you can make the amp apparently saturate, as the capacitors run out of voltage at LF.
Other possible cheap (or not so cheap) hacks include adding an autoformer to create your own CT, or using a large series inductor, to get the winding's DC voltage, filtering the ~35V ripple on it.
Tim
If an amplifier doesn't draw much current through the CT use 2 adequate filter capacitors in series with the centre connected to your voltage divider. The filter capacitors will be supplying the music waveform at the lowest frequency rather than the transformer as in CT power supplies. A simple voltage divider will work fine and the amplifier doesn't really care if the supply rails get a bit offset from the signal.
This will not work with DC coupled amplifiers, so plan on rolling off the low end response at 20Hz or so depending o the size of the filter capacitors.
A couple of power darlingtons and appropriate zener diodes can be used to clamp the virtual ground and extend the frequency down to DC if necessary, most of the time they will dissipate no heat only conducting if the ground drifts too far from centre.
This will not work with DC coupled amplifiers, so plan on rolling off the low end response at 20Hz or so depending o the size of the filter capacitors.
A couple of power darlingtons and appropriate zener diodes can be used to clamp the virtual ground and extend the frequency down to DC if necessary, most of the time they will dissipate no heat only conducting if the ground drifts too far from centre.
Which circuit you use depends on if you have a ready made circuit board or not both circuits do the same thing and have the same low frequency roll off, the single supply design is using the amplifier in bit more conventional manner in that the differential input doesn't see much common mode voltage at low frequency unlike the voltage divider split supply.
If you bridge two amplifiers for each channel, you could run them from a single supply and don't need any output capacitors. This works well with a 'fake' ground as described above, because there is no return ground current directly from the speaker. The artificial ground is only used as an input reference, and won't require much current, which is what you want.
Search for BTL or bridged gainclone and you should find something on this site.
Search for BTL or bridged gainclone and you should find something on this site.
If the transformer core isn't welded, you can unwind it and add a centre tap. I've done that more than once with Radio Shack transformers. That was a long time ago when I was younger and stupider, so I don't remember precisely how I determined where to put the CT. Remove all the secondary, mark the middle of the wire, then rewind? Unwind a bunch of turns, then test the voltage and calculate how many more to take off or add? Count all the turns, then wind back 1/2 and put the CT there? I wouldn't have split the secondary wire and wound it back bifilar -while that should guarantee matched windings, it doesn't seem as safe due to the much larger voltage difference in adjacent wires.
Have a look at some articles on transformer winding before you tackle this. It's not rocket surgery. I got all I needed to know from the ARRL Handbook and a chapter in the "Electricity 11" correspondence course.
Have a look at some articles on transformer winding before you tackle this. It's not rocket surgery. I got all I needed to know from the ARRL Handbook and a chapter in the "Electricity 11" correspondence course.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.