OK don't shoot me, I've read a lot of posts on this subject and I'm in need of clarification. I'm not an EE, not by any means, so please be gentle!
1. Does the input impedance of my amp6 change with the position of my volume pot? I thought it must do but then how can you match it?
2. Can I simply change the output impedance of my modified PS1 by changing the bridging resistor at the outputs?
3. Does anyone know how to easily measure these things?
Hope you can help chaps.
Lee
1. Does the input impedance of my amp6 change with the position of my volume pot? I thought it must do but then how can you match it?
2. Can I simply change the output impedance of my modified PS1 by changing the bridging resistor at the outputs?
3. Does anyone know how to easily measure these things?
Hope you can help chaps.
Lee
(1). Probably by less than 10%.
(2). No.
(3). Yes, but you need some kit.
Your input impedance is probably the same as that of the volume control, and will be written on the side of the volume control. If it's a transistory thing it will probably be 10k. If it's new valve it will probably be 100k. If it's old valve it might be 500k.
Output impedance is probably harder. If it has op-amps then it's probably less than 100 Ohm. If it has discrete transistors, then it's probably less than 600 Ohm. If it's valve, it could be anything. New valve might be less than 2k Ohm, old valve could be up to 50k Ohm.
Why do you need to know?
(2). No.
(3). Yes, but you need some kit.
Your input impedance is probably the same as that of the volume control, and will be written on the side of the volume control. If it's a transistory thing it will probably be 10k. If it's new valve it will probably be 100k. If it's old valve it might be 500k.
Output impedance is probably harder. If it has op-amps then it's probably less than 100 Ohm. If it has discrete transistors, then it's probably less than 600 Ohm. If it's valve, it could be anything. New valve might be less than 2k Ohm, old valve could be up to 50k Ohm.
Why do you need to know?
You do not need to worry about impedance matching with line level audio signals. Most devices are designed to have low output impedance and high input impedance, i.e. the outputs "look" like voltage sources to the high resistance/impedance inputs.
Yes the input impedance of the amp varies somewhat with the volume control setting, but unless you are using really long cables, or driving the input from a high impedance source (there are a very few around), you don't have to worry about it.
Phonograph inputs that are driven by relatively high impedance magnetic cartridges do require special attention to get proper frequency response.
I_F
Yes the input impedance of the amp varies somewhat with the volume control setting, but unless you are using really long cables, or driving the input from a high impedance source (there are a very few around), you don't have to worry about it.
Phonograph inputs that are driven by relatively high impedance magnetic cartridges do require special attention to get proper frequency response.
I_F
Cool, so the current set-up is OK? I only listen at very low levels so the pot is only about 30% over most of the time.
Forgive the sketch quality....not exactly a schematic but it should make sense.
Is there anything you would change?
The T-amp has a 25K input resistor......and 25K feedback if that makes any difference.
edit: The mod is as per the "PS1 as a CDP " thread on the digital forum, wired straight from the DAC through a DC blocking cap to the RCA's with a 100k resistor to ground.
Also, does the pot then alter the corner frequency of the RC filter?
Forgive the sketch quality....not exactly a schematic but it should make sense.
Is there anything you would change?
The T-amp has a 25K input resistor......and 25K feedback if that makes any difference.
edit: The mod is as per the "PS1 as a CDP " thread on the digital forum, wired straight from the DAC through a DC blocking cap to the RCA's with a 100k resistor to ground.
Also, does the pot then alter the corner frequency of the RC filter?
Attachments
Hi,
is Tamp wired as inverting?
Has Tamp a DC blocking cap on it's input?
Yes, the following stage (cables and resistors and parasitic capacitance) will affect the turn over frequency of the RC output filter in the DAC.
Any loading less than infinite will have an effect.
Choose components that make the effect inaudible.
From your schematic, it appears you have 100k (DAC resistor) // 25k (input impedance) // 25k (pot-guess) = 11k1.
The turn over frequency (just due to the resistors) is 3Hz rather than the 0.3Hz due to 100k alone.
If there is a second DC block then that effect will aslo have to be calculated.
is Tamp wired as inverting?
Has Tamp a DC blocking cap on it's input?
Yes, the following stage (cables and resistors and parasitic capacitance) will affect the turn over frequency of the RC output filter in the DAC.
Any loading less than infinite will have an effect.
Choose components that make the effect inaudible.
From your schematic, it appears you have 100k (DAC resistor) // 25k (input impedance) // 25k (pot-guess) = 11k1.
The turn over frequency (just due to the resistors) is 3Hz rather than the 0.3Hz due to 100k alone.
If there is a second DC block then that effect will aslo have to be calculated.
AndrewT said:
OK so I can change the pot to 50K that's not a problem, I have one.
I do currently have another set of DCblockers in the T-amp but I was going to integrate th elot together and just have one set before the pot...or maybe after???
I can change the DAC resistor as well if needed, what do you think?
Hi,
a DC block on each item makes them universally compatible with other equipment.
For the above reason I would keep both sets of cap blockers.
You should now calculate what bass roll-off frequency you require and what components can achieve it.
Two equal value series connected caps have an effective capacitance equal to half a single cap.
For different value caps the combination can be calculated using the same technique as parallel resistors.
I would change the DAC to 10uF with 1M0 as the grounding resistor.
I would also put 10uF in the amp with a 2M2 grounding resistor before the cap. These grounding resistors pull any cap leakage to virtually zero volts and can sometimes permit live disconnection and reconnection. Although the preferred method is to always switch off, particularly since some amplifiers can become unstable when there is no source impedance setting the input conditions.
I generally recommend a bass filter roll-off frequency one decade below the lowest usable audio frequency you intend reproducing i.e. about -3db @ 2Hz and -1db @ 4Hz, these reduce phase and reponse errors to virtual inaudibility.
However there is another criterion for the input filter. It should be at least half an octave above the NFB loop bass roll-off frequency and this in turn should be at least another half octave above the PSU roll-off frequency. This rule is generally ignored by many and seems to cause few problems. If you decide to adopt it then it demands a PSU -3db of no higher than 1Hz for an input filter of 2Hz. Some will say that the caps needed to give this will sound less good than smaller caps chosen for good mid range and treble. Experience and trial listening may be your solution.
a DC block on each item makes them universally compatible with other equipment.
For the above reason I would keep both sets of cap blockers.
You should now calculate what bass roll-off frequency you require and what components can achieve it.
Two equal value series connected caps have an effective capacitance equal to half a single cap.
For different value caps the combination can be calculated using the same technique as parallel resistors.
I would change the DAC to 10uF with 1M0 as the grounding resistor.
I would also put 10uF in the amp with a 2M2 grounding resistor before the cap. These grounding resistors pull any cap leakage to virtually zero volts and can sometimes permit live disconnection and reconnection. Although the preferred method is to always switch off, particularly since some amplifiers can become unstable when there is no source impedance setting the input conditions.
I generally recommend a bass filter roll-off frequency one decade below the lowest usable audio frequency you intend reproducing i.e. about -3db @ 2Hz and -1db @ 4Hz, these reduce phase and reponse errors to virtual inaudibility.
However there is another criterion for the input filter. It should be at least half an octave above the NFB loop bass roll-off frequency and this in turn should be at least another half octave above the PSU roll-off frequency. This rule is generally ignored by many and seems to cause few problems. If you decide to adopt it then it demands a PSU -3db of no higher than 1Hz for an input filter of 2Hz. Some will say that the caps needed to give this will sound less good than smaller caps chosen for good mid range and treble. Experience and trial listening may be your solution.
AndrewT said:
Wow, now things are getting really interesting...and harder to follow....
Nevertheless...I'm thinking of putting the whole thing in one new box so I only need i set of caps (4.7uf should do OK now, yes?. I have some really nice Poly caps and would hate to waste them). I've been told that DC damages the slide contact on a potentiometer so I'll put them before it.
Now for the resistors...Ohms I know, K/Ohms also but what's an M? is it a Mega-Ohm!!!!
As for the PSU, I have a 10,000uf stiffener off board.
Crikey this has taken on a whole new angle for me, I can build, solder and follow directions without fault...I even learned how these Tripath chips function...but this impedance thing has got me in a real spin!!....maybe I should go back to college...no maybe not!
Hi,
keep reading the Forum, it continues to teach me a lot.
M=mega 10^6
k = kilo 10^3
m=milli 10^-3
u~=micro 10^-6 (u should be the greek mu but I don't know how to get it on screen)
But please don't write 10,000 uF when 10mF means the same. Oh dear, that doesn't help, since the Continentals this side of the pond use" ," to mean decimal place not thousands separator.
Except some manufacturers mix m and M, u and m, just to confuse us.
keep reading the Forum, it continues to teach me a lot.
M=mega 10^6
k = kilo 10^3
m=milli 10^-3
u~=micro 10^-6 (u should be the greek mu but I don't know how to get it on screen)
But please don't write 10,000 uF when 10mF means the same. Oh dear, that doesn't help, since the Continentals this side of the pond use" ," to mean decimal place not thousands separator.
Except some manufacturers mix m and M, u and m, just to confuse us.
AndrewT said:
Debatable, I would say. Common engineering and scientific practice would be to use mF instead of 1000 uF. However, by tradition in electronics, capacitances of mF sizes are usually written as thousands of uF, and since some idi.... write mF when they mean uF, maybe it is least confusing to actually stick to the 1000 uF practice.
As for potential confusion, how about this page?
http://www.mcs.surrey.ac.uk/Personal/R.Knott/Fibonacci/propsOfPhi.html#phiBrick
They use both pi, phi and Phi, all written out in letters, instead of writing the greek letters. I haven't checked if they get it right, but I would surprised if they managed without mixing them up somewhere on the page.
Why does this forum not let us use correct symbols?
I have always thought ASCI table were international. If i try to write micro (alt + 230) here: µ
I can clearly see it correct while editing post as seen in atachment
Bute not after posting (at least not last time I tried), why?
edit:
Ok, now it worked! Fine then
edit2:
ohmega: Ù
edit 3:
No this does not work
I have always thought ASCI table were international. If i try to write micro (alt + 230) here: µ
I can clearly see it correct while editing post as seen in atachment
Bute not after posting (at least not last time I tried), why?
edit:
Ok, now it worked! Fine then
edit2:
ohmega: Ù
edit 3:
No this does not work
Attachments
I searched for omega, but is seems not to be part of the western ASCII, so no chance...
IF you have windows, goto system programs, there is a program listing all chars for selected fonts (don't know english name), select as font "verdana", disable unicode and select western code.
These are the chars available.
Mike
Edit: You can also use "Alt Gr"+m, the normal way to get a "µ"... Or Strg+Alt+m.
IF you have windows, goto system programs, there is a program listing all chars for selected fonts (don't know english name), select as font "verdana", disable unicode and select western code.
These are the chars available.
Mike
Edit: You can also use "Alt Gr"+m, the normal way to get a "µ"... Or Strg+Alt+m.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.