Hi,
I thought it might be helpful to write a couple of comments for my PA03 build (My First attempt at Audio electronics)
Firstly as a note of caution - I followed the 'Farnell' Bill of Materials from the download.
The order numbers for the Resistors led me to 500mW (1/2W) 1% tolerance manufactured by Multicomp.
These particular Resistors are WIDER than the PCB's footprint that is printed on the board.
I struggled to get the multiple resistor banks to fall within the PCB footprint - I had to solder resistor 1 and 3 in place (for example) then solder resistor 2 Above it's neighbours. The Resistors around the IC sockets were particularly troublesome.
Pavel's PCB design has the elegance of ' Meager sufficiency ' - But this doesn't allow the use of components with physical dimensions different from the one intended.
Welwyn Resistors from farnell - 500mW 1% tolerance Are the correct size
I've seen that people use the term 'stuffing' for the populating of a PCB with components.
Very Apt for my board - It looks well and truly Stuffed!!
I thought it might be helpful to write a couple of comments for my PA03 build (My First attempt at Audio electronics)
Firstly as a note of caution - I followed the 'Farnell' Bill of Materials from the download.
The order numbers for the Resistors led me to 500mW (1/2W) 1% tolerance manufactured by Multicomp.
These particular Resistors are WIDER than the PCB's footprint that is printed on the board.
I struggled to get the multiple resistor banks to fall within the PCB footprint - I had to solder resistor 1 and 3 in place (for example) then solder resistor 2 Above it's neighbours. The Resistors around the IC sockets were particularly troublesome.
Pavel's PCB design has the elegance of ' Meager sufficiency ' - But this doesn't allow the use of components with physical dimensions different from the one intended.
Welwyn Resistors from farnell - 500mW 1% tolerance Are the correct size
I've seen that people use the term 'stuffing' for the populating of a PCB with components.
Very Apt for my board - It looks well and truly Stuffed!!
Very useful information for me.
I am collecting parts for this project.
Hope that I can start to build it with in weeks.
I am collecting parts for this project.
Hope that I can start to build it with in weeks.
I would like to have a better understanding of the significance of input/output impedance and how it effects the source-amp-speaker path.
Why do some systems 'like' to 'see' low impedance, others high?
Maybe someone could direct me to an existing source of information (I know google is my friend ;-) it's also very non-specific) or can take the time to reply
Thanks in advance
Why do some systems 'like' to 'see' low impedance, others high?
Maybe someone could direct me to an existing source of information (I know google is my friend ;-) it's also very non-specific) or can take the time to reply
Thanks in advance
most of our audio equipment uses voltage to transfer signals between modules.
For good voltage transfer one adopts low source impedance and high receive impedance.
eg.
the output impedance of a cdp could be 200ohms.
The input impedance of your passive volume control could be 50k
The ratio of 200:50k gives a very high transfer of the available voltage.
It is normally recommended that the ratio for adequate voltage transfer be 1:5, good transfer uses 1:10 and excellent uses >1:20
But there is a secondary requirement.
The connection between modules must be able to pass the audio signal. If we are working in analogue then the audio pass band is 20Hz to 20kHz.
To pass this without any roll off and with low phase shift one usually arranges for the F-3dB roll offs to occurs outside the range of 2Hz to 200kHz.
The impedances of the source and receiver affect the passband.
Use the formula F-3dB = 1/ {2 Pi R C} for a passive RC filter.
eg.
200ohms source impedance with a cable of 100pF/m that is 5m long to a remote Receiver that has a 10k input impedance with a 47pF RF filter on it's input. The CDP has 470nF 250V polypropylene film and foil DC blocking cap on it's output.
The cable and the RF filter add up to a capacitive load of 547pF
The low pass filter formed by the 200ohms and the 547pF gives an F-3dB of 1.4MHz
The 200ohms source impedance has ensured that the capacitive loading does not roll off the treble end of the audio passband.
The 470nF capacitor has to drive the 10k vol pot.
The F-3dB is 34Hz.
This is more than a decade higher than needed to pass all of the Low Frequency end of the audio passband.
Two solutions. Leave as is and accept the rolled off bass, or change the vol pot to 50k and change the DC blocking capacitor to 2u2F 63V Mylar. New F-3dB ~1.4Hz
This will perform better than the very expensive polypropylene adopted by the cdp manufacturer.
For good voltage transfer one adopts low source impedance and high receive impedance.
eg.
the output impedance of a cdp could be 200ohms.
The input impedance of your passive volume control could be 50k
The ratio of 200:50k gives a very high transfer of the available voltage.
It is normally recommended that the ratio for adequate voltage transfer be 1:5, good transfer uses 1:10 and excellent uses >1:20
But there is a secondary requirement.
The connection between modules must be able to pass the audio signal. If we are working in analogue then the audio pass band is 20Hz to 20kHz.
To pass this without any roll off and with low phase shift one usually arranges for the F-3dB roll offs to occurs outside the range of 2Hz to 200kHz.
The impedances of the source and receiver affect the passband.
Use the formula F-3dB = 1/ {2 Pi R C} for a passive RC filter.
eg.
200ohms source impedance with a cable of 100pF/m that is 5m long to a remote Receiver that has a 10k input impedance with a 47pF RF filter on it's input. The CDP has 470nF 250V polypropylene film and foil DC blocking cap on it's output.
The cable and the RF filter add up to a capacitive load of 547pF
The low pass filter formed by the 200ohms and the 547pF gives an F-3dB of 1.4MHz
The 200ohms source impedance has ensured that the capacitive loading does not roll off the treble end of the audio passband.
The 470nF capacitor has to drive the 10k vol pot.
The F-3dB is 34Hz.
This is more than a decade higher than needed to pass all of the Low Frequency end of the audio passband.
Two solutions. Leave as is and accept the rolled off bass, or change the vol pot to 50k and change the DC blocking capacitor to 2u2F 63V Mylar. New F-3dB ~1.4Hz
This will perform better than the very expensive polypropylene adopted by the cdp manufacturer.
Last edited:
Yes I see.
So, when implementing an Amp/Preamp design, maximum power exchange is dictated by how close to Ideal one matches source impedance to receiver's, Without losing carried signal quality.
And such a solution is Unique to every Source-Cable-Receiver example.
I can actually do a better job than the manufacturers since I can adjust for my specific implementation.
Correct?
So, when implementing an Amp/Preamp design, maximum power exchange is dictated by how close to Ideal one matches source impedance to receiver's, Without losing carried signal quality.
And such a solution is Unique to every Source-Cable-Receiver example.
I can actually do a better job than the manufacturers since I can adjust for my specific implementation.
Correct?
NOT power exchange.
Voltage transfer.
For general voltage transfer efficiency one adopts a low output impedance and a high input impedance. And checks for bandwidth limitations due to parasitic and intended capacitances.
An example of doing it right is to fit the DC blocking capacitor in the Power Amplifier input and select the components for the RC filter effect to suit the frequencies you require the Power Amplifier to pass to the speaker.
The manufacturers of Source equipment take a guess at what could be connected by the majority of users/operators and then decide to fit components that suit that majority for the "quality" they think helps sell their product.
For maximal power exchange the rules are very different. RF reception requires maximal power transfer. Here you must match the impedance of the source to the receiver.
eg. for a radio/TV tuner the aerial (source), the cable and input impedance of the Receiver all need to have the same impedances. Same for a video very wideband signal, a 75ohm video connection requires the source to have a 75ohms output impedance, the connectors and cable to have a 75ohms characteristic impedance and the receiver to have a 75ohms impedance. But this time it's not so much for power transfer but rather for minimising reflection of signals that will corrupt the wanted signal received.
Last edited:
"NOT power exchange"
Sorry, You stated in your original explanation that source 'Ideal' is low impedance and receiver 'Ideal' is high.
my fault was in incorrect choice of words . my words imply i was describing 'Impedance Matching' of RF.
Just one more follow on question -
You state
"An example of doing it right is to fit the DC blocking capacitor in the Power Amplifier input and select the components for the RC filter effect to suit the frequencies you require the Power Amplifier to pass to the speaker."
I have made some loudspeaker designs and the passive crossovers have always been placed at the speaker input. Do you mean that Xover is placed at amp input?
My intention is to use Two PA03 amp boards (1 x 120W for Sub + 2x60W 2-way monitors) for a 'Hifi' (ish) PC based sound system.
Would it therefore be more efficient to place the 1st order low-pass filter at the Amp input rather than Speaker input?
AND if so could higher order filters also be placed at amp input? - Is there any increased efficiency? - Obviously the amp becomes a 'One Trick Pony' tied to the specific speaker the Xover was designed for.
( I appreciate you taking the time to answer my 'electronics 101' queries. )
Sorry, You stated in your original explanation that source 'Ideal' is low impedance and receiver 'Ideal' is high.
my fault was in incorrect choice of words . my words imply i was describing 'Impedance Matching' of RF.
Just one more follow on question -
You state
"An example of doing it right is to fit the DC blocking capacitor in the Power Amplifier input and select the components for the RC filter effect to suit the frequencies you require the Power Amplifier to pass to the speaker."
I have made some loudspeaker designs and the passive crossovers have always been placed at the speaker input. Do you mean that Xover is placed at amp input?
My intention is to use Two PA03 amp boards (1 x 120W for Sub + 2x60W 2-way monitors) for a 'Hifi' (ish) PC based sound system.
Would it therefore be more efficient to place the 1st order low-pass filter at the Amp input rather than Speaker input?
AND if so could higher order filters also be placed at amp input? - Is there any increased efficiency? - Obviously the amp becomes a 'One Trick Pony' tied to the specific speaker the Xover was designed for.
( I appreciate you taking the time to answer my 'electronics 101' queries. )
I learned something very important tonight. The wiser folks amongst you will already know this but I will share anyway as it could save those less experienced folks a lot of time and worry.
If you are using a bulb tester, the output protection circuit will not work consistently. If all your voltages check out, but the protection circuit is still wacky, plug it in without the bulb tester. Instant joy!
This took me far too long to figure out.
I now have two PA03 that sound as beautiful as they look. One gave me lots of grief, one worked perfectly upon assembly. Thank you to upupa epops and peranders for making this learning experience possible.
Fingers crossed in the hopes that my first order will arrive soon.
If you are using a bulb tester, the output protection circuit will not work consistently. If all your voltages check out, but the protection circuit is still wacky, plug it in without the bulb tester. Instant joy!
This took me far too long to figure out.
I now have two PA03 that sound as beautiful as they look. One gave me lots of grief, one worked perfectly upon assembly. Thank you to upupa epops and peranders for making this learning experience possible.
Fingers crossed in the hopes that my first order will arrive soon.