I've run across this topology in Rockford Fosgate car amplifiers, specifically the Punch 500. The "output" capacitors are also the supply's filter capacitors. If you note the output's collectors are grounded while the emitters are connected to the rails. The resulting "swing" is actually a modulation of the power supply. Since DC is the lowest frequency the amplifier will ever (rather, never) reproduce, using the capacitors in this form is actually ingenious. Crazy, but it works and works well. The amplifier is inherently self-protecting, to a degree, against short circuits and is capable of driving very low impedances without placing undue stress on the outputs. Inductive spiking is reduced to a minimum because of the capacitive "cushion" the filter caps provide.
My brother runs QSC amplifiers for PA work and once had an MX700 running into a dead short for a 2 hour period at full power. It got hot but didn't cause any damage!
My brother runs QSC amplifiers for PA work and once had an MX700 running into a dead short for a 2 hour period at full power. It got hot but didn't cause any damage!
(A bit late to enter this thread)
Peranders wrote:
I was also kind of surprised when I saw an output grounded amp for the first time. The topology has one big disadvantage (according to a skilled designer), the amp has little rejection of noise from the mains if you don't have a very advanced (shielded) transformer.
Then Nelson Pass:
Naaah, there's nothing special about the PSRR of the
topology either way.
...
To my understanding this is not a question of PSRR, but the fact that the power supply is not grounded. Therefore any capacitive leakage current is injected in the amplifier and creates a voltage across the feedback resistor. I use this topology and had some difficulties at the beginning with this fact.
Peranders wrote:
I was also kind of surprised when I saw an output grounded amp for the first time. The topology has one big disadvantage (according to a skilled designer), the amp has little rejection of noise from the mains if you don't have a very advanced (shielded) transformer.
Then Nelson Pass:
Naaah, there's nothing special about the PSRR of the
topology either way.
...
To my understanding this is not a question of PSRR, but the fact that the power supply is not grounded. Therefore any capacitive leakage current is injected in the amplifier and creates a voltage across the feedback resistor. I use this topology and had some difficulties at the beginning with this fact.
they both share is that you can't easily run multiple channels from one psu. Or did someone already say that? I didn't read the thread thoroughly...
I am actually also using the circlotron topology. I reinvented it (as many others, I see) by wanting to have a fully symmetrical circuit using only NMOS fets and by shifting the + and - power supplies around.
There was a lot of power supply noise at the beginning, but now it is nearly gone (even without using a shield winding in the transformers). I don't know exactly why; probably because I changed the feedback resistors to low values (4k7).
There was a lot of power supply noise at the beginning, but now it is nearly gone (even without using a shield winding in the transformers). I don't know exactly why; probably because I changed the feedback resistors to low values (4k7).
Hi; the first time I saw this kind of configuration: collectors of final transistors grounded and the positive terminal for speaker taken from the central point of the main capacitors, I got surprised. I built the amplifier QSC1700 following the schematic carefully and it sounds good but I found its gain was low; I made some adjustments to the resistors in the input opamp circuitry. This amplifier does not use DC detect because of the coupled capacitors, isn´t it?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.