I had typed up about 10 minutes worth of blurb and then the power went off here at work, so I'm not going to re-do it. 
This weekend I might try the circlotron cct as per the attached schematic. After that? I might just try and cook something up along the lines of one of Nelson Pass' X amps. That should get tongues wagging hey? Of course it's all just for fun, just messing around. Not selling any. Really, truly, Mr. Pass' lawyer...
GP.
This weekend I might try the circlotron cct as per the attached schematic.
After that? I might just try and cook something up along the lines of one of Nelson Pass' X amps. That should get tongues wagging hey? Of course it's all just for fun, just messing around. Not selling any. Really, truly, Mr. Pass' lawyer... GP.
Attachments
Way too conventional.
Hi Jan. Good to hear from you. Actually the other night I did check out some LM395 data. But basically I don't have any in the junk box, and I do have some '317s and this afternoon some LM337's too. I dont need to make an amplifier at all really, and if I did there are better ways of doing it. This is something I thought about doing more than 25 years ago and now I finally am doing something about it. Can't hurry these things you know. Anyway, one thing I am going to have a play with is a complementary symmetry setup with a 317 on top and a 337 on the bottom. With a bipolar, or especially a fet setup when you ease the bias voltage on, the standing current goes up gradually with the bias. Not so here, the current will go from zero to max once the upper and lower ic's start pulling against each other, thanks to the gain of the internal opamps. So what I will try is to set the bias just below the point at which the ic's start their conduction tug-of-war i.e. true class B. A millivolt or so of crossover distortion may or may not be noticeable. Big experiment. If that is a total failure I will put a resistor of some low value resistor between the ic outputs and dial up the bias to several milliamps. Have a cap from each ic output pin to the loudspeaker. One interesting thing is that once the bias is set there should be no need for thermal feedback of any sort - it should just stay where it is set. I am hoping for about 9 watts into 8 ohms with a +/- 18v supply. The limit is 1.5 amps peak. This is fun.
GP.
Hi Jan. Good to hear from you. Actually the other night I did check out some LM395 data. But basically I don't have any in the junk box, and I do have some '317s and this afternoon some LM337's too. I dont need to make an amplifier at all really, and if I did there are better ways of doing it. This is something I thought about doing more than 25 years ago and now I finally am doing something about it. Can't hurry these things you know.
Anyway, one thing I am going to have a play with is a complementary symmetry setup with a 317 on top and a 337 on the bottom. With a bipolar, or especially a fet setup when you ease the bias voltage on, the standing current goes up gradually with the bias. Not so here, the current will go from zero to max once the upper and lower ic's start pulling against each other, thanks to the gain of the internal opamps. So what I will try is to set the bias just below the point at which the ic's start their conduction tug-of-war i.e. true class B. A millivolt or so of crossover distortion may or may not be noticeable. Big experiment. If that is a total failure I will put a resistor of some low value resistor between the ic outputs and dial up the bias to several milliamps. Have a cap from each ic output pin to the loudspeaker. One interesting thing is that once the bias is set there should be no need for thermal feedback of any sort - it should just stay where it is set. I am hoping for about 9 watts into 8 ohms with a +/- 18v supply. The limit is 1.5 amps peak. This is fun. GP.
Definitely is fun!
Yes, most of the time we are too impatient to wait for the delivery delays.
Pity you don't have LM395's, although they are "super transistors" with only a few uA base current for 1A Ic, they do exhibit the typical transistor turn-on behaviour of a diode junction. Would ease your bias setting.
How about offset? Are you not getting a DC offset across the speaker? You can use the coupling caps, but a little opamp to balance the DC....?
Jan Didden
Yes, most of the time we are too impatient to wait for the delivery delays.
Pity you don't have LM395's, although they are "super transistors" with only a few uA base current for 1A Ic, they do exhibit the typical transistor turn-on behaviour of a diode junction. Would ease your bias setting.
How about offset? Are you not getting a DC offset across the speaker? You can use the coupling caps, but a little opamp to balance the DC....?
Jan Didden
Re: OK. Here's the rules ;-)
How about an amplifier made from a TDA8172/TDA8177 Vertical Deflection amplifier IC from an old monitor?
Datasheet
I talked a friend into trying it earlier tonight,and it works!
Hopefully we'll have a schematic and stuff soon..
Here's a link to a thread on Geek's forum about it:
off the wall amp
Next we're thinking of playing around with LM317/337's,while we're at it,so I had to find this old thread.
Is there any new news on the LM317/337 amp?
Circlotron said:1/ Any amplifiers built must use parts not meant for audio usage, or at least as few as possible .
2/ The more unusual (or silly) the topology the better.
3/ Preferably use junk box parts, the lower the cost the better.
4/ Don't be afraid to think outside the box. e.g. 4049UB like Daniel, or how about a CRT for an amplifying stage?
How about an amplifier made from a TDA8172/TDA8177 Vertical Deflection amplifier IC from an old monitor?
Datasheet
I talked a friend into trying it earlier tonight,and it works!
Hopefully we'll have a schematic and stuff soon..
Here's a link to a thread on Geek's forum about it:
off the wall amp
Next we're thinking of playing around with LM317/337's,while we're at it,so I had to find this old thread.
Is there any new news on the LM317/337 amp?
Jeff Macaulay headphone amp circuit
Hallo folks,
This post is a reply to the message written by Samuel Jayaraj. It is true that mr.Macaulay once published a headphone amp circuit based on LM317 regulator i.c.'s. What's more, I have built a stereo version of that circuit and it sounded wonderful (until I hooked up the power supply with inverse polarity, which was the end both of my amp and of a cheap pair of cans... ).
Here is the complete article reference:
J.Macaulay, "Irregular use of regulators in headphone amplifier", Electronics and Wireless World, October, 856, 1997
For the truly curious I attach a scan of the circuit.
By the way, if anyone is looking for fleapower: Build an amp using TL431s rather than LM317s (it is quite possible).
Keep your soldering irons heated! Best regards from rainy Holland,
arensattic
Hallo folks,
This post is a reply to the message written by Samuel Jayaraj. It is true that mr.Macaulay once published a headphone amp circuit based on LM317 regulator i.c.'s. What's more, I have built a stereo version of that circuit and it sounded wonderful (until I hooked up the power supply with inverse polarity, which was the end both of my amp and of a cheap pair of cans...
).Here is the complete article reference:
J.Macaulay, "Irregular use of regulators in headphone amplifier", Electronics and Wireless World, October, 856, 1997
For the truly curious I attach a scan of the circuit.
By the way, if anyone is looking for fleapower: Build an amp using TL431s rather than LM317s (it is quite possible).
Keep your soldering irons heated! Best regards from rainy Holland,
arensattic
Attachments
Higher Powered LM317 amp version
Hallo Dr.EM,
Yes, it could be possible to build a higher-powered version of the Macaulay circuit. I have never built such a version myself, but John Broskie once suggested a 4W SE version in his "Tube CAD Journal".
For your convenience, I add a scan of John's circuit to this post.
Best regards from VERY rainy and VERY windy Holland (Beaufort 9 and several centimeters of water in the streets....)
Arensattic
Hallo Dr.EM,
Yes, it could be possible to build a higher-powered version of the Macaulay circuit. I have never built such a version myself, but John Broskie once suggested a 4W SE version in his "Tube CAD Journal".
For your convenience, I add a scan of John's circuit to this post.
Best regards from VERY rainy and VERY windy Holland (Beaufort 9 and several centimeters of water in the streets....)
Arensattic
Attachments
Well, it works . I made essentially the circuit arensattic posted but it is run single supply, capacitive coupled (artificial ground point made with op-amp voltage follower, planar tweeters will need cap anyhow). Sure gets toasty with that fairly small heatsink, a fan currently keeps it happy but that is one amp of two; on the same sink . Not got much listening in but it does sound decent from what I've heard. Really could do with more power dupply filtering but humming won't get to the tweeters anyhow. Scope naturally showed clean sine wave output.
. I made essentially the circuit arensattic posted but it is run single supply, capacitive coupled (artificial ground point made with op-amp voltage follower, planar tweeters will need cap anyhow). Sure gets toasty with that fairly small heatsink, a fan currently keeps it happy but that is one amp of two; on the same sink . Not got much listening in but it does sound decent from what I've heard. Really could do with more power dupply filtering but humming won't get to the tweeters anyhow. Scope naturally showed clean sine wave output.Attachments
OK, so I'm 9 years late to this party.... put the music is still playing!
I came across this circuit by chance and it peaked my interest for two reasons.....
1) I've always wanted to try a Class-A amp.
2) I'm too lazy and dumb to build discrete.
I'm surprised more people haven't experimented with this circuit. If you do any DIY you probably have of the components on hand already.
I put together a split supply version using an OPA134 to add some gain. The current source is set for 1.14A, which makes the whole thing get pretty toasty when I switch on the +/-16V supply. Ouput offset is only 7mV!
The circuit sounds suprisingly good playing through a large speaker. I had the current source set lower originally but the bass sounded thin and uncontrolled even at low volume. This has the lowest noise floor of any amp I have built so far though. I can press my ear right against the tweeter and can't even tell it's powered on.
I have some LT1083 7.5A adjustable regs here and I think I might try out a higher powered version. Winter is right around the corner anyway.
I came across this circuit by chance and it peaked my interest for two reasons.....
1) I've always wanted to try a Class-A amp.
2) I'm too lazy and dumb to build discrete.
I'm surprised more people haven't experimented with this circuit. If you do any DIY you probably have of the components on hand already.
I put together a split supply version using an OPA134 to add some gain. The current source is set for 1.14A, which makes the whole thing get pretty toasty when I switch on the +/-16V supply. Ouput offset is only 7mV!
The circuit sounds suprisingly good playing through a large speaker. I had the current source set lower originally but the bass sounded thin and uncontrolled even at low volume. This has the lowest noise floor of any amp I have built so far though. I can press my ear right against the tweeter and can't even tell it's powered on.
I have some LT1083 7.5A adjustable regs here and I think I might try out a higher powered version. Winter is right around the corner anyway.
I came across this circuit by chance and it peaked my interest for two reasons.....
1) I've always wanted to try a Class-A amp.
2) I'm too lazy and dumb to build discrete.
I'm surprised more people haven't experimented with this circuit. If you do any DIY you probably have of the components on hand already.
I put together a split supply version using an OPA134 to add some gain. The current source is set for 1.14A, which makes the whole thing get pretty toasty when I switch on the +/-16V supply. Ouput offset is only 7mV!
The circuit sounds suprisingly good playing through a large speaker. I had the current source set lower originally but the bass sounded thin and uncontrolled even at low volume. This has the lowest noise floor of any amp I have built so far though. I can press my ear right against the tweeter and can't even tell it's powered on.
I have some LT1083 7.5A adjustable regs here and I think I might try out a higher powered version. Winter is right around the corner anyway.
Hi anonymous
Interesting, do you have a drawing of the OPA + regulators circuit to share ?
Circlotron is undoubtedly the forefather of regulators chip amps.I developed independently my own strains.
Some bear an obvious resemblance, others don't.
Have a look here (and follow the sequence of links):
http://www.diyaudio.com/forums/chip-amps/193214-class-chip-amp-now-complementary-version.html
Those amplifiers do have shortcomings, but they can be surprisingly good and effective: I really used some of them for everyday listening, and they were very satisfying.
It's the 3rd image down on this page.....
index
I found your thread after trying to find other examples of regulator based amps. Yours is undoubtedly superior to this basic circuit, but with 3x opamps and 3x regs, it took the novel simplicity away for me.
I'll post some pics after I put together my "monster" LT1083 version (if it even works). Need to find a suitable power supply though.
Here is the LM317 version I have been joyfully listening to for the past few hours (opa134 is under the resistors)....
Attachments
This is a later version, and remember, this BOM applies to a complete stereo amplifier.
http://www.diyaudio.com/forums/chip-amps/192934-se-class-regulator-chip-amp-madness.html
This one is also stereo, and uses 2 opamps and 3 regs, that is one and a half reg + one opamp per channel: less than your own version in fact.
And the interesting thing about this circuit is that although it remains truly class A and single-ended, it also has an improved efficiency.
Nice circuit Elvee, I think I might try that one for headphones.
I want to built a "power amp" version with LT1083. Wouldn't the shared current source limit my output power? Maybe 7.5A is enough to share for two channels since we can only go so high with the voltage anyway (plus that is some serious heat).
BTW, what do you think a max safe rail voltage would be? Would +/-18V be cutting it to close to the device limits?
EDIT: I see the LT1083 has an input to output limit of only 30V instead of 40V like the LM317. Bummer. Might have to go bridged.
I want to built a "power amp" version with LT1083. Wouldn't the shared current source limit my output power? Maybe 7.5A is enough to share for two channels since we can only go so high with the voltage anyway (plus that is some serious heat).
BTW, what do you think a max safe rail voltage would be? Would +/-18V be cutting it to close to the device limits?
EDIT: I see the LT1083 has an input to output limit of only 30V instead of 40V like the LM317. Bummer. Might have to go bridged.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.