Hi,
the DC block can be electrolytic or non-electrolytic.
The N-E types can be metalised film or foil and insulator or ceramic, or silvered mica etc. These all have very low leakage. i.e. when a voltage is applied across the capacitors the resistance is VERY large and the current that passes (rather than charging) is tiny (nA, pA, fA). However all the electrolytic types have a much lower resistance to the applied voltage. Some leakage current flows. This leakage current will be measurable with most domestic setup and even with our low quality instruments.
If the output has a resistance to ground of a few tens of ohms and the DC blocking capacitor has something of the order of 1M0 of RESISTANCE then you have a very effective resistance divider.
Apply 20V to the series combination of 1M0 and 10r and measure the voltage across the 10r. ~0.2mV can be measured with a reasonable DMM with a high input impedance and a sensitive voltage scale.
Now disconnect the 10r output resistor and the air across the terminals becomes a resistor with a value that could be around 1G0 (1000M). Calculate the voltage across the terminals now with that pair of resistors forming the dividing ladder. Don't bother trying to measure it accurately. The input resistance of the DMM will completely distort the resistor ratios. What you will be able to measure is the resistor ladder of the 1M0 of the blocking capacitor and the ~10M of the DMM. I guess you will get a reading that looks very like 20V.
Does that make any sense?
Now go and look at some circuits that have DC blocking caps on the inputs and outputs.
Do any of them have a high value resistor between the signal pin and ground. It appears to be useless. Guess what that resistor is doing?
the DC block can be electrolytic or non-electrolytic.
The N-E types can be metalised film or foil and insulator or ceramic, or silvered mica etc. These all have very low leakage. i.e. when a voltage is applied across the capacitors the resistance is VERY large and the current that passes (rather than charging) is tiny (nA, pA, fA). However all the electrolytic types have a much lower resistance to the applied voltage. Some leakage current flows. This leakage current will be measurable with most domestic setup and even with our low quality instruments.
If the output has a resistance to ground of a few tens of ohms and the DC blocking capacitor has something of the order of 1M0 of RESISTANCE then you have a very effective resistance divider.
Apply 20V to the series combination of 1M0 and 10r and measure the voltage across the 10r. ~0.2mV can be measured with a reasonable DMM with a high input impedance and a sensitive voltage scale.
Now disconnect the 10r output resistor and the air across the terminals becomes a resistor with a value that could be around 1G0 (1000M). Calculate the voltage across the terminals now with that pair of resistors forming the dividing ladder. Don't bother trying to measure it accurately. The input resistance of the DMM will completely distort the resistor ratios. What you will be able to measure is the resistor ladder of the 1M0 of the blocking capacitor and the ~10M of the DMM. I guess you will get a reading that looks very like 20V.
Does that make any sense?
Now go and look at some circuits that have DC blocking caps on the inputs and outputs.
Do any of them have a high value resistor between the signal pin and ground. It appears to be useless. Guess what that resistor is doing?
Caps pass DC until charged. If there is no connection to ground, it won't charge. A voltmeter presents a high resistance to ground so the cap won't charge and you'll read a DC offset.
With the speaker connected, there is now a low resistance to ground, and after the initial charge (which you'll hear as a small thump in your speaker).. you will read close to 0V of DC on the speaker terminal. This is correct.
Dual supplies are usually used on power amps to eliminate this output capacitor - this minimises the distortion introduced by the capacitor, and also helps to minimise the turn on thud.
With the speaker connected, there is now a low resistance to ground, and after the initial charge (which you'll hear as a small thump in your speaker).. you will read close to 0V of DC on the speaker terminal. This is correct.
Dual supplies are usually used on power amps to eliminate this output capacitor - this minimises the distortion introduced by the capacitor, and also helps to minimise the turn on thud.
Thanks for explaining things I have observed...
From my experiments on headphone amps that did require DC blocking on the output, I learned that, the lower the impendance of the load, the larger that cap needs to be to pass bass frequencies... for 8 and 4 ohm loads, I would easily suggest 10000uf there...
From my experiments on headphone amps that did require DC blocking on the output, I learned that, the lower the impendance of the load, the larger that cap needs to be to pass bass frequencies... for 8 and 4 ohm loads, I would easily suggest 10000uf there...
Yes, Nuuk.... Sue, Grabbit and Run, Sols are still alive and well in Southern England !
Thanks everyone for your help. I think I can finish it off now in the knowledge that it is safe.
On to the next project. An envelope dropped on to my mat today from Chipamp.com an LM3886 based amp. One for the winter nights. $2.3 dollars to the £
he he
Thanks everyone for your help. I think I can finish it off now in the knowledge that it is safe.
On to the next project. An envelope dropped on to my mat today from Chipamp.com an LM3886 based amp. One for the winter nights. $2.3 dollars to the £
he he- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.