Why no 10K option in your list? 🙂
Seriously though, it's a question that's unanswerable without context. What are the characteristics of the signal source?
Seriously though, it's a question that's unanswerable without context. What are the characteristics of the signal source?
As low as possible to reject hum.
As high as possible so as not to load pre amp/mixer.
10K is good. I have one class d amp that is 3k3 input impedance.
On the other hand I have a valve amp that is 1Meg input impedance.
As high as possible so as not to load pre amp/mixer.
10K is good. I have one class d amp that is 3k3 input impedance.
On the other hand I have a valve amp that is 1Meg input impedance.
Not necessarily...higher impedance can cause an increase in noise. As always, there are trade-offs to consider regarding best performance for a given situation. Most MM phono cartridges are loaded with 47 kohms while a MC phono cart will usually need to be loaded with a few tens or hundreds of ohms. Consumer line level inputs are frequently spec'd at 10 kohms, but anything from a few kohms to 100 kohms is valid depending on application. If your cable runs aren't too long, say a couple meters or so, then a lower input impedance level will theoretically be less noisy while not adversely affecting frequency response. Longer cable runs will require higher impedance loading to avoid premature high frequency roll off. The output impedance is also relevant here, although most well designed modern circuits can generally be safely assumed to have an output impedance of around 150 ohms or less, and capable of driving almost any load it's likely see.
Mike
Last edited:
Certainly it can if you listen to an amp or preamp with no source connected. But who does that? In operation the input impedance is shunted by the output impedance of the driving stage which is typically several orders of magnitude lower in value.
Well, again, not necessarily...If the input impedance is set to a highish value on a negative feedback type (pre)amp then the feedback network will usually need to reflect that to maintain a reasonable DC offset, and many opamps need to see relatively equal source impedance on their inputs to keep CM distortion low as possible.
Every application has it's own requirements, blanket statements claiming this or that without the relevant contextual information are often not true.
Mike
Every application has it's own requirements, blanket statements claiming this or that without the relevant contextual information are often not true.
Mike
Sure its quite possible that with a pathological design, noise is going to be an issue with a higher input impedance. But I was assuming relatively competent design.
If you have a question it would help if you'd spell it out explicitly. That is if you're here to learn.
OK...I'll play.
Both of the circuits below have a gain of 6dBv (X2). The upper one has an input impedance of 50k, and the lower one has an input impedance of a little less than 2k. The higher impedance one will have slightly higher noise and a little more CM distortion than the one below.
So is high impedance still always better?
Mike
Absolutely nothing. You are free to use what ever circuit topology you wish, I was just using those to illustrate my point.
Mike
Well your circuit is your responsibility, if it performs below the standards you aimed at, how do you intend to correct it?
I merely note that the CM distortion is going to be a function not of the input impedance resistor chosen but rather the AC impedance seen by the opamp with the signal source in circuit. What's the output impedance of that signal source you have?
I merely note that the CM distortion is going to be a function not of the input impedance resistor chosen but rather the AC impedance seen by the opamp with the signal source in circuit. What's the output impedance of that signal source you have?
If you refer to the OPA2134 data sheet: http://www.ti.com/lit/ds/symlink/opa134.pdf - sec 7.3.3 "Source Impedance and Distortion" you'll see that the low impedance version meets those requirements without regard to source impedance, and it works just fine. 😉
Come to think of it, using lower circuit impedance levels also helps to minimize the effects of stray circuit capacitance. Who knew! 😀
Mike
No answer to my question then about the source impedance you've chosen? The paragraph you've cited begins :
For lowest distortion with a source or feedback network with an impedance greater than 2 kΩ
You think >2kohm for a source impedance is likely in this context?
For lowest distortion with a source or feedback network with an impedance greater than 2 kΩ
You think >2kohm for a source impedance is likely in this context?
Take another look at the circuit, the non-inverting input sees 2kohm at DC and less than that for AC signal. Inverting input sees 2k, 4k in parallel with 4k. Both inputs satisfy the requirements as laid out in the data sheet, therefore CM distortion will be minimized regardless of source impedance.
Mike
So why choose such high feedback resistors in your first example circuit (post #13)? Because you care more about DC than audio frequencies?
- Status
- Not open for further replies.