An amp I am working on has a preamp/tone control like the pic below. The output is fed to a differential pair, not shown, obviously...
I am going to add pre-out.main-in jacks, and plan on breaking the signal at the output of this circuit. However, it appears to me that the drive capability of this circuit will be pretty poor, so I planned on adding a OPA2604 as a buffer between the two stages.
Honestly never used a opamp as a simple buffer (unity gain). A friend emailed me and said to install a resistor in the feedback loop equivilent to the output impedence of the circuit feeding the amp, although he didn't say what this circuit might require. (10K parallelled with 27K....somewhere around 7K? Or feedback resistor not needed?) Shooting for a reasonable degree of DC precision...
Any thoughts on simple buffer circuits and perhaps the best way to incorporate one into this?
I am going to add pre-out.main-in jacks, and plan on breaking the signal at the output of this circuit. However, it appears to me that the drive capability of this circuit will be pretty poor, so I planned on adding a OPA2604 as a buffer between the two stages.
Honestly never used a opamp as a simple buffer (unity gain). A friend emailed me and said to install a resistor in the feedback loop equivilent to the output impedence of the circuit feeding the amp, although he didn't say what this circuit might require. (10K parallelled with 27K....somewhere around 7K? Or feedback resistor not needed?) Shooting for a reasonable degree of DC precision...
Any thoughts on simple buffer circuits and perhaps the best way to incorporate one into this?
Attachments
Download the datasheet of BUF634. At the end you have the simplified schematic of the circuit (Diamond buffer) or you could also check my QRV-03 headphone amp and the output stage. The korean guy below is very satisfied with it.
You could also check Walter Jung's article.
http://www.elecdesign.com/Globals/PlanetEE/Content/2800.html
http://www.headphoneamp.co.kr/bbs/v...&sc=on&select_arrange=headnum&desc=asc&no=604
http://www4.head-fi.org/forums/showthread.php?s=&threadid=35074
You could also check Walter Jung's article.
http://www.elecdesign.com/Globals/PlanetEE/Content/2800.html
http://www.headphoneamp.co.kr/bbs/v...&sc=on&select_arrange=headnum&desc=asc&no=604
http://www4.head-fi.org/forums/showthread.php?s=&threadid=35074
Yep, I've read most of that at one time or another.
The BUF634 is a fine chip, but way overkill for what I want. Plus it means ordering more parts from a supplier where the minimum order is $25 US and the BUF634 is about $7 US each, and I'd need two. I have several OPA2604's just sitting here, and I'd like to put one of them to use. They would work fine for my purposes in this mid-fi amplifier. I was just hoping to gain some insight into using an audio opamp such as the 2604 in a unity gain application.
So far, the only suggestion I've gotten is the resistor in the feedback loop. Is this adviseable? What value?
Any other ideas?
The BUF634 is a fine chip, but way overkill for what I want. Plus it means ordering more parts from a supplier where the minimum order is $25 US and the BUF634 is about $7 US each, and I'd need two. I have several OPA2604's just sitting here, and I'd like to put one of them to use. They would work fine for my purposes in this mid-fi amplifier. I was just hoping to gain some insight into using an audio opamp such as the 2604 in a unity gain application.
So far, the only suggestion I've gotten is the resistor in the feedback loop. Is this adviseable? What value?
Any other ideas?
Does your circuit work? Have you calculated the bias for each stage? This 5 pF? Do you need this?
I think driving capability for the tone control network seems to be low. Also the wire between base and emitter (= short circuit), what the purpose, error maybe? If you want a very simple buffer, just use an emitter follower.
I think driving capability for the tone control network seems to be low. Also the wire between base and emitter (= short circuit), what the purpose, error maybe? If you want a very simple buffer, just use an emitter follower.
I don't know where that wire came from....between the base of the second transistor and the 150K resistor...pls ignore, looks like a drawing artifact of some sort.
This is a fully functional gain stage/tone amp on a commercial product. Not terribly high-end, but the integrated amp here was bought to utilize the amplifier section, which is quite good sounding. So, to use the amp section I need to add the preout/main-in jacks. The buffer is not necessary for the main-in section, and I won't use the preout much, but I can see that unless I add some sort of buffer to the output of this circuit that the preamp section will be nearly useless, and I'd like it to be functional. Hence the questions about using a simple opamp such as the 2604 as a buffer.
This is a fully functional gain stage/tone amp on a commercial product. Not terribly high-end, but the integrated amp here was bought to utilize the amplifier section, which is quite good sounding. So, to use the amp section I need to add the preout/main-in jacks. The buffer is not necessary for the main-in section, and I won't use the preout much, but I can see that unless I add some sort of buffer to the output of this circuit that the preamp section will be nearly useless, and I'd like it to be functional. Hence the questions about using a simple opamp such as the 2604 as a buffer.
Here are three basic unity gain circuits. The third is an example of what my friend told me...to make the resistor the approximate value of the output impedence of the supplying stage.
What are the pluses and minuses of using any one of these?
You replies are really appreciated. Thank you much...
What are the pluses and minuses of using any one of these?
You replies are really appreciated. Thank you much...
Attachments
One or two...
Hi EchoWars...
The first exemple in your figure doesn´t have unity gain ,but the gain of two..
Regards
Hi EchoWars...
The first exemple in your figure doesn´t have unity gain ,but the gain of two..
Regards
Standard application is "no resistor". It will work with no massive problems. Data sheets for op amps will say, to be picky, match impedances on op amp inputs IF the series resistance on the (+) input exceeds a certain value. For the OPA 2134 this value is 2k. For the OPA 2604 I don't know.
In your case: if this is an emitter follower buffer, in other words, zero or just a few dozen ohms between preceding stage output and the (+) input, then the feedback path can be a straight wire. If there is a series resistance exceeding say, these 2k, then make the feedback path from out to (-)in a resistor of equal value to the (+)in series resistance.
The main effect will be in lowering DC offset.
MBK
In your case: if this is an emitter follower buffer, in other words, zero or just a few dozen ohms between preceding stage output and the (+) input, then the feedback path can be a straight wire. If there is a series resistance exceeding say, these 2k, then make the feedback path from out to (-)in a resistor of equal value to the (+)in series resistance.
The main effect will be in lowering DC offset.
MBK
OR:
use an inverting buffer: the circuit as is to (-)in, a 27k//10k from -(in) to output, and (+)in to GND.
Usually overshoot and capacitive drive capability as well as DC offset problems are lesser in inverting configuration.
MBK
use an inverting buffer: the circuit as is to (-)in, a 27k//10k from -(in) to output, and (+)in to GND.
Usually overshoot and capacitive drive capability as well as DC offset problems are lesser in inverting configuration.
MBK
Re: One or two...
MBK, the datasheet on the 2604 does not indicate at what point the series resistance should be matched, but in the circuit in my first post, it appears that the output impedence is somewhere around 10K. This seems like a large value for a feedback resistor in a unity gain stage. If I have examined the supplying stage incorrectly (certainly possible, since I'm obviously rusty on this), I'd appreciate being corrected.
I cannot invert. There is no latter stage where I might invert again, so I'm stuck with non-inverting. Perhaps I may stick a trimpot in the feedback loop and see what it does to offset, which is my main consideration...from there I might read a value and decide on a resistor...
Yes.....saw my error after posting, was too lazy to change.Tube_Dude said:
MBK, the datasheet on the 2604 does not indicate at what point the series resistance should be matched, but in the circuit in my first post, it appears that the output impedence is somewhere around 10K. This seems like a large value for a feedback resistor in a unity gain stage. If I have examined the supplying stage incorrectly (certainly possible, since I'm obviously rusty on this), I'd appreciate being corrected.
I cannot invert. There is no latter stage where I might invert again, so I'm stuck with non-inverting. Perhaps I may stick a trimpot in the feedback loop and see what it does to offset, which is my main consideration...from there I might read a value and decide on a resistor...
It will really not matter much whether you use straight wire or a 10k resistor. Think of the "typical" volume control in most active preamps: it is a 50k or 100k pot to gnd with pot wiper to the (+)in of a buffer. The series impedance so created in front of a non-inverting buffer configuration is variable and can be high. It is not a 100% "clean" topology to me... yet it bothers very few designers. In fact I haven't heard a single comment on this even though you can read about impedance matching of = and - inputs in many datasheets.
Me, I'd still put a 10k in the feedback path... but wherever I did that in my buffers it just didn't seem to matter much for the end result. Much more important for the performance of your buffer will be proper op amp power supply decoupling (see your datasheet), and to add a series R of 47 to 200 Ohms to the output to make it able to drive cables (a capacitive load).
MBK
Me, I'd still put a 10k in the feedback path... but wherever I did that in my buffers it just didn't seem to matter much for the end result. Much more important for the performance of your buffer will be proper op amp power supply decoupling (see your datasheet), and to add a series R of 47 to 200 Ohms to the output to make it able to drive cables (a capacitive load).
MBK
Thank you very much for your input. I'm hoping for decent DC offset at the output (<5mV). I've been successful in obtaining this level of offset in the past with the OPA2604 being used as a gain stage, but I have a feeling that if used as a unity-gain buffer that I will not be so lucky.
Considered getting a few OPA604's, which have offset adjust pins, but I'm shooting for simplicity as well as DC accuracy...might be a battle to get both.
The supply voltage is +/- 24V, which is why I want to use the 2604. I like its sound, and I don't need to order parts. If I can keep offset low, then I'll consider this project a success.
Considered getting a few OPA604's, which have offset adjust pins, but I'm shooting for simplicity as well as DC accuracy...might be a battle to get both.
The supply voltage is +/- 24V, which is why I want to use the 2604. I like its sound, and I don't need to order parts. If I can keep offset low, then I'll consider this project a success.
For what it's worth, in my preamp and active X-0 using OPA 2134 I get that range of DC offset (5-10 mV) ... ... and that after a series of maybe 10 op amps. I don't even use SPICE or such, just impedance matching of the "good enough" kind and using inverters where possible. With 2604 it should work just fine as well.
MBK
MBK
Thank you. 🙂
Amp still on its way from somewhere in Canada, so I'll get to fool with it next week if I'm lucky. But with your help it appears I may be prepared.
Amp still on its way from somewhere in Canada, so I'll get to fool with it next week if I'm lucky. But with your help it appears I may be prepared.
Sorry to resurrect this old thread:
What do you guys think of this circuit?
Will it sound decent?
That's because I'm having one of those Alps 6 channel motorized pot and would like to make a simple 6 channel pre-amp to be fed from a modified Behringer DCX2496 that has passive output. The 6 channel pre will be feeding to a 6 channel Rotel amp having input impedance of about 22K.
So you see there is a need for some active input buffering (from a passive output of the DCX) and then driving into a 22K load.
What do you guys think of this circuit?
Will it sound decent?
That's because I'm having one of those Alps 6 channel motorized pot and would like to make a simple 6 channel pre-amp to be fed from a modified Behringer DCX2496 that has passive output. The 6 channel pre will be feeding to a 6 channel Rotel amp having input impedance of about 22K.
So you see there is a need for some active input buffering (from a passive output of the DCX) and then driving into a 22K load.
Attachments
- Status
- Not open for further replies.