|16th February 2011, 05:31 PM||#12|
likes 1's more than 0's
instead of the voltage being measured/referenced against ground '0v' the second wire that would be signal ground is actively driven in the opposite polarity by a matching inverted amplifier, so the headphone cable needs to have 4 wires, 1 for each phase per channel. so often the only reference to ground is in the power supply. then at the driver any errors that exist in both phases/wires are cancelled out (to the degree that the common mode rejection ratio (CMRR) of the circuit/device permits) when they are added together. that can be distortion, noise, interference etc. purists object that also some good stuff like 2nd harmonic distortion, which is pleasant to the ear, is also cancelled out and makes it sound sterile, i disagree.
so as a result you end up with a transmission line that has double the voltage swing (only if the power supply is doubled to drive the second phase), very low distortion and very good immunity to noise. i find that low level detail is preserved to a higher degree and that bass has more slam, an overall cleaner sound to these ears. then you also have the pass SuSy designs but you can search the pass forum for that, this is what is called a balanced single ended amplifier and that is the type of circuit that the opa1632 opamp in the board i posted uses.
as a result, with my portable headphone rig, not only does it sound brilliant, but i'm never bothered by that horrible cell phone interference when someone takes a call or sends a text, its cancelled out
hope that helps
Last edited by qusp; 16th February 2011 at 05:34 PM.
|23rd February 2011, 06:46 PM||#13|
Join Date: Dec 2010
All the schematics you posted are class A/B output.
So all use the NPN/PNP complimentary resistors.
The main difference I see is the amount of capacitors and filtering by the "added" circuitry of the more complex looking designs.
The other important difference I see is the amount of diodes...and one uses an LED (light emitting diode- which I add the def just in case).
The diodes cause current and voltage to continually keep the transistors active so you never have one fully off. This reduces crossover dstortion.
One uses 2 diodes, one uses one diode the other uses an LED. I have had success using all combos. So play around to find out what you prefer...Class B can sound suprisingly good if sensitive high speed transistors are used. Class B is when the transistor closes completely until enough voltage opens it...this ommits the need for the diodes and the two bases of the BJT's are connetced with nothing in between.
I have done a circuit for larger 2 way speakers- 10"woofer x 1"tweeter- 90db. I have driven this speaker directly from op amps. LF353 and TL082 were able to drive the speakers louder than you would like sitttin' with your ear to 1" from the speaker. So smaller transistors will only help and larger power transistors are not necessary- something that can be heatsinked will work fine...unless you do pure class B...then less heat is generated.
Just get a decent hfe and audio grade. Darligntons are ok...though I prefer straight BJT- richer smoother sound- at least I think. I've had the best sound with TIP41/42. MJ200/ 210 give decent sound and are smaller. I've driven smaller speakers with 2n3904 and 3906.
Another great sounding combo is NTE11/12- even in class B. They are high current and have a max ic of 5amps, but a low power dissapation.
(I know NTE is hated- but I haven't found the cross reference) Class B will give off very little heat. Just do a straight wire connecting the 2 bases of the BJT's.
I also keep my opamp designs very low gain. 1K - 1K....5k -5k on the feedback resistor/ gain adjusting resistor. Using +/-12 with a 2Vrms input results in 10V pk/pk. So I only need to amplify current This leaves me with 8.5V pk/pk after going throu the output BJT stage using .7ohm resistors on the emitters. This max volume results in zero clipping.
With an 8 ohm speaker I get 6.6V rms, 818mA rms wich translates into 5.39W rms. So I need to be sure everything can handle peak to peak ratings of 17v, 2.1amps. Then everything will be safe at full volume and will result in zero clipping. Class B will result in very little heat.
One thing that is important is the resistors on the on the emitters. This will help prevent thermal runaway should the resistors overheat. Though the risk is low at such low power, the last thing you want is to destroy your cans...especially when wearing them.
Last edited by rcollege; 23rd February 2011 at 06:49 PM. Reason: clarification
|Thread Tools||Search this Thread|
|Thread||Thread Starter||Forum||Replies||Last Post|
|2x Gain pure op-amp headphone amp idea||Dxvideo||Headphone Systems||30||10th September 2011 03:05 AM|
|Poor power output, op amp based headphone amp.||dannybeckett||Solid State||13||27th September 2010 09:31 PM|
|Headphone Amp:njm4556s Op Amp For Rs1 And Others||uhcmos||Headphone Systems||7||19th October 2008 09:02 AM|
|your ADI op-amp suggestions, please||dfeweer||Parts||2||3rd August 2008 11:09 PM|
|Headphone amp with power OP?||Zen Mod||Headphone Systems||2||9th June 2005 03:15 AM|
|New To Site?||Need Help?|