Well, I always wanted to choke load a source-follower, and now I find myself building a headphone amp for my 32Ohm grados, and I thought I might as well do just that. Ok, my question is, if I use a IRF510 with a single suply of 12V (running of a large battery I have) and would like a bias current of say 500mA, how do I go about choosing what choke values I need? Do I just choose a value of mHs that doesn't roll-off frequencies with an impedance of 32 ohms? And, as for the resistance, this will set the bias current, correct? Now, do I choose a choke, and then set the voltage a the source to have the desired current run through whatever resistance the choke is? If so, then wouldn't I just be throwing away voltage if the source isn't at 1/2 the suply voltage? Otherwise, do I choose a choke with a resistance so that at 1/2 the suply voltage there will be say 500mA running through it? I'm not too sure on what are the importantfactors of choke loading a source-follower, so any other help would be apreciated. I'm not too sure on how to wind my own high value chokes, but I can buy hammond chokes close to here, if that makes a difference.
Hey JoeBob,
What is the reason for choking the output? Is this to change the frequency resonce? Create a filter? If so, I would add a much smaller filter at the input of the amp to get rid of the frequencies you don't want. Why amplify frequencies that you are going to get rid of anyway? This would be less effiecient.
I understand that you want to use a 12V battery to power this amp. A class A amp circuit would not be very efficient and would reduce the time that the battery will have charge. If this is a headphone amp, you wouldn't want to have much power...even 1W would probably do some hearing damage when placed right in your ear. I would use a class AB capacitor coupled follower for the output stage, run off of 12V. This would have a MUCH lower bias current saving your battery. If you want to use descrete componants (I would) then use complementary componants for output follower stage. These would not have to be very large transistors because this is a low power circuit and a bias current of a few milliamps would be perfect. Choose small Lo-noise devices for the differential amp stage and current source\current mirror. A Lo-noise device would be good for the VAS stage too as I am assuming you want to get highest quality with lowest possible noise for this project. Is there any particular reason you want to use IRF510 for the amp? Might want to check the data sheets for this Mosfet as it may not be that linear of a device.
Driving a 32Ohm load at this small power would not require much current at all, and with 12V a class AB follower circuit of less than 1W would not produce much heat. You might be able to use TO-92 case style devices for the output stage. Or maybe TO-126 case style transistors W\out heat sink would be better because you would be operating more in the linear region of the devices.
Even if you made a class A output stage from one 12V battery, you would have to capacitor couple(or transformer couple) to the 32Ohm speaker in order to block the DC or else your headphone would
.
I have included a descrete amp circuit topology that may work well for this application and have QUALITY responce. You will have to figure out the bias resistors and experiment, but this illistrates my point. Just include a passive filter circuit for an input filter and choose the RC componants to give you the frequency cuttoff that you desire.
BTW: You could use surface mount transistors(SOT-23) for all in this circuit except the 2 output transistors(TO-126), depending on your soldering skills. 🙄
What is the reason for choking the output? Is this to change the frequency resonce? Create a filter? If so, I would add a much smaller filter at the input of the amp to get rid of the frequencies you don't want. Why amplify frequencies that you are going to get rid of anyway? This would be less effiecient.
I understand that you want to use a 12V battery to power this amp. A class A amp circuit would not be very efficient and would reduce the time that the battery will have charge. If this is a headphone amp, you wouldn't want to have much power...even 1W would probably do some hearing damage when placed right in your ear. I would use a class AB capacitor coupled follower for the output stage, run off of 12V. This would have a MUCH lower bias current saving your battery. If you want to use descrete componants (I would) then use complementary componants for output follower stage. These would not have to be very large transistors because this is a low power circuit and a bias current of a few milliamps would be perfect. Choose small Lo-noise devices for the differential amp stage and current source\current mirror. A Lo-noise device would be good for the VAS stage too as I am assuming you want to get highest quality with lowest possible noise for this project. Is there any particular reason you want to use IRF510 for the amp? Might want to check the data sheets for this Mosfet as it may not be that linear of a device.
Driving a 32Ohm load at this small power would not require much current at all, and with 12V a class AB follower circuit of less than 1W would not produce much heat. You might be able to use TO-92 case style devices for the output stage. Or maybe TO-126 case style transistors W\out heat sink would be better because you would be operating more in the linear region of the devices.
Even if you made a class A output stage from one 12V battery, you would have to capacitor couple(or transformer couple) to the 32Ohm speaker in order to block the DC or else your headphone would
. I have included a descrete amp circuit topology that may work well for this application and have QUALITY responce. You will have to figure out the bias resistors and experiment, but this illistrates my point. Just include a passive filter circuit for an input filter and choose the RC componants to give you the frequency cuttoff that you desire.
BTW: You could use surface mount transistors(SOT-23) for all in this circuit except the 2 output transistors(TO-126), depending on your soldering skills. 🙄
Attachments
Well, as for choke loading the source-follower, it's not to filter out frequencies, it's because a choke loaded source-follower is more efficient than a resistor loaded class-A follower, and I think it's more efficient that a ccs loaded one as well. Why class-A, why 500mA? Because I like the class-A sound, I like it alot, and my experience with class-A poweramps, and the IRF510 is that they sound better running hot. 500mA might be overkill, half that is more than enough, but more can never hurt (500mA at 12V won't put out too much heat for a TO-220 package). And I know I'll be needing an output cap.
It'll be batery powered from a 4.5Ah battery, so I think I'll be able to play for a good 4 hours straight, I'll set it up to recharge while powered off. A simple Class-AB circuit, with say complimentary BJTs would work well, I could even drive the load with opamps most likely, but I like the sound of a class-a source follower, it's just my preferance.
I just don't know much about using a choke instead of a source resistor or ccs, I'm not saying it'll sound better, I just want to try it.
It'll be batery powered from a 4.5Ah battery, so I think I'll be able to play for a good 4 hours straight, I'll set it up to recharge while powered off. A simple Class-AB circuit, with say complimentary BJTs would work well, I could even drive the load with opamps most likely, but I like the sound of a class-a source follower, it's just my preferance.
I just don't know much about using a choke instead of a source resistor or ccs, I'm not saying it'll sound better, I just want to try it.
- Status
- Not open for further replies.