Hello,
I am refurbishing a SS preamp and have a couple of questions about a portion of the power amp output circuit. This particular preamp has 2 sets of outputs, and neither the service manual or the user manual indicate that there is any difference between them. Also, the output impedance of both is specified as 47ohms.
Here are my questions:
1) R279/R280 are both 47ohm resistors. I recognize that impedance is not simply resistance, but are these resistors used to specify the output impedance of the preamp? Perhaps a better question: what sets the output impedance in this circuit?
2) R607/R608 (both 10k) are used on the L/R of OUT1, but not OUT2 - wouldn't that suggest that the 2 outputs would behave differently? As I mentioned above, the specifications do not indicate any differences between these outputs. What is the purpose of R607/R608?
3) Am I correct to assume that the purpose of C175/C176 are used for decoupling i.e. blocking DC from the output?
What's interesting about all of this is that, though specified in the circuit, the board itself has R607/R608 & C175/C176 "hacked" in i.e. there are no locations for these parts on the board. I suspect that I have an early production run of the board, or something like that, and they were omitted in error.
Thanks in advance for any help provided!
Bob
I am refurbishing a SS preamp and have a couple of questions about a portion of the power amp output circuit. This particular preamp has 2 sets of outputs, and neither the service manual or the user manual indicate that there is any difference between them. Also, the output impedance of both is specified as 47ohms.
Here are my questions:
1) R279/R280 are both 47ohm resistors. I recognize that impedance is not simply resistance, but are these resistors used to specify the output impedance of the preamp? Perhaps a better question: what sets the output impedance in this circuit?
2) R607/R608 (both 10k) are used on the L/R of OUT1, but not OUT2 - wouldn't that suggest that the 2 outputs would behave differently? As I mentioned above, the specifications do not indicate any differences between these outputs. What is the purpose of R607/R608?
3) Am I correct to assume that the purpose of C175/C176 are used for decoupling i.e. blocking DC from the output?
What's interesting about all of this is that, though specified in the circuit, the board itself has R607/R608 & C175/C176 "hacked" in i.e. there are no locations for these parts on the board. I suspect that I have an early production run of the board, or something like that, and they were omitted in error.
Thanks in advance for any help provided!
Bob
Attachments
Those resistors will dictate the output impedance if we assume that the driving source (opamp ?) has essentially zero ohms output impedance.
The resistors will provide a measure of protection in the event of a short being applied.
The 10uF cap does provide DC isolation. It also influences the output impedance at low frequencies because of the reactance of the cap. In practice, and working into say a 100k input impedance of the following equipment there is no problem.
R607/608 define a zero volts condition on the cap output, something that is always good practice.
The resistors will provide a measure of protection in the event of a short being applied.
The 10uF cap does provide DC isolation. It also influences the output impedance at low frequencies because of the reactance of the cap. In practice, and working into say a 100k input impedance of the following equipment there is no problem.
R607/608 define a zero volts condition on the cap output, something that is always good practice.
Thanks for your reply, Mooly!
I see now that I wasn't reading the schematic correctly with regards to R607/608. Your response helped clear that up for me. Thanks again!
I see now that I wasn't reading the schematic correctly with regards to R607/608. Your response helped clear that up for me. Thanks again!
The 47 ohm resistors there are to isolate the output circuitry from the external load capacitance which may cause the output amplifier to oscillate and also reduces the likelihood of external RF interference being rectified [detected] and amplified..
(Once upon a very long time ago I designed consumer, pro and MI electronics for a living)
(Once upon a very long time ago I designed consumer, pro and MI electronics for a living)
@rgbeach, can you tell us the make and model number of those so we can have a closer look...?
you are right about the output impedance as not just being 47 ohms, it depends on the topology, most ic op amps i know have about 600 ohm output impedance, so it depends on the circuit really..
you are right about the output impedance as not just being 47 ohms, it depends on the topology, most ic op amps i know have about 600 ohm output impedance, so it depends on the circuit really..
Not really.
Ideal Op Amp output impedance is zero, real World ones may show a couple ohms, if that much.
yes, my bad....the 5532 has less than an ohm of output resistance...https://www.onsemi.com/pub/Collateral/NE5532-D.PDF
they tested using a 600 ohm load...
they tested using a 600 ohm load...
Let's clear this up.
SHORTS: the '709 opamp was rated for few-second shorts. They were awful easy to kill when poking. Essentially all later chips are protected for indefinite shorts. The 47r series resistor is not required for short protection.
Output Impedance: many opamps have emitter-follower outputs at ~~1mA so will approach tens-of-Ohms internal Zout. Some have more. A few have much more.
But we never listen to opamps open-loop. In use, Zout is divided by the excess gain thrown-away in NFB.
NE5532: datasheet says 0.3r at 10KHz with closed-loop gain 30dB (32). Open-loop gain at 10KHz is 70dB (3,000). Assume intrinsic Zout is 30 Ohms. The 40dB(100) of NFB turns this into 0.3 Ohms.
TL072: Unlike many chip amps, this has about 160 Ohms of dumb resistance in series with the output pin. Open loop bandwidth is 3MHz, so at 10KHz the gain is 300. Working unity-gain Zout is 160/300 or a half-Ohm. Working at gain of 32 (which is a lot for '072) Zout @10KHz is like 16 Ohms.
Driving Cables: Zout is dominated by excess gain which falls with frequency. The '072 at unity gain is 0.05r at 1KHz, 0.5r @10KHz, 5r @100KHz, and 50r@1MHz. What does this remind us of? An inductor, about 8uH. OTOH the cable looks like a capacitor, about 300pFd for 30 feet (10m). We have a (emulated) coil and a capacitor. Radio-fans know this as a Tuned Circuit, which can be excited at some specific frequency. For these values, around 3MHz. If there is "no" pure resistance, the excitations can build up very big. For these values a 47 Ohm resistor limits the Q to about 3, which is not real high as these things go. A value of 330r makes Q=0.5 which is very tame.
This resistor goes OUTside the NFB loop. Effectively at the start of the external cable. Resistance inside the NFB has no good dynamic effect. (It will reduce chip dissipation when delivering large currents; hence the ~160r inside the '072.)
LOAD: while the small-signal output impedance is often near one Ohm, the opamp will not drive much signal into few-Ohm loads. For many many chips you can assume a 10mA current limit (to avoid '709-style sudden death). If you also assume a 10V peak swing, you see that you "can" drive a 1K load pretty big. A fair number of chips are rated for more. Many will make sine-like shapes in 600r, even to 13V (22mA) peak.
But loading-down any amplifier cuts its gain. NFB keeps closed-loop gain constant, but we lose NFB's ability to reduce THD. '5532 was designed for heavy loading and does very well in 600r, does not suck at 300r. '072 was designed to be cheap. It covers loads over 2K extremely well, but THD degrades below 2K.
SHORTS: the '709 opamp was rated for few-second shorts. They were awful easy to kill when poking. Essentially all later chips are protected for indefinite shorts. The 47r series resistor is not required for short protection.
Output Impedance: many opamps have emitter-follower outputs at ~~1mA so will approach tens-of-Ohms internal Zout. Some have more. A few have much more.
But we never listen to opamps open-loop. In use, Zout is divided by the excess gain thrown-away in NFB.
NE5532: datasheet says 0.3r at 10KHz with closed-loop gain 30dB (32). Open-loop gain at 10KHz is 70dB (3,000). Assume intrinsic Zout is 30 Ohms. The 40dB(100) of NFB turns this into 0.3 Ohms.
TL072: Unlike many chip amps, this has about 160 Ohms of dumb resistance in series with the output pin. Open loop bandwidth is 3MHz, so at 10KHz the gain is 300. Working unity-gain Zout is 160/300 or a half-Ohm. Working at gain of 32 (which is a lot for '072) Zout @10KHz is like 16 Ohms.
Driving Cables: Zout is dominated by excess gain which falls with frequency. The '072 at unity gain is 0.05r at 1KHz, 0.5r @10KHz, 5r @100KHz, and 50r@1MHz. What does this remind us of? An inductor, about 8uH. OTOH the cable looks like a capacitor, about 300pFd for 30 feet (10m). We have a (emulated) coil and a capacitor. Radio-fans know this as a Tuned Circuit, which can be excited at some specific frequency. For these values, around 3MHz. If there is "no" pure resistance, the excitations can build up very big. For these values a 47 Ohm resistor limits the Q to about 3, which is not real high as these things go. A value of 330r makes Q=0.5 which is very tame.
This resistor goes OUTside the NFB loop. Effectively at the start of the external cable. Resistance inside the NFB has no good dynamic effect. (It will reduce chip dissipation when delivering large currents; hence the ~160r inside the '072.)
LOAD: while the small-signal output impedance is often near one Ohm, the opamp will not drive much signal into few-Ohm loads. For many many chips you can assume a 10mA current limit (to avoid '709-style sudden death). If you also assume a 10V peak swing, you see that you "can" drive a 1K load pretty big. A fair number of chips are rated for more. Many will make sine-like shapes in 600r, even to 13V (22mA) peak.
But loading-down any amplifier cuts its gain. NFB keeps closed-loop gain constant, but we lose NFB's ability to reduce THD. '5532 was designed for heavy loading and does very well in 600r, does not suck at 300r. '072 was designed to be cheap. It covers loads over 2K extremely well, but THD degrades below 2K.
thanks, i was able to download but my eyes are old and can not see well the very small drawings...
in the past, i would email the manufacturer with my technical queries stating the model number and serial number of the unit, sometimes i get lucky and got a reply...
blocking caps at the output of the op-amps are done to eliminate the possible damage it may cause to the power amp it feeds if the power amp is direct coupled at the input..
in most cases may not be needed, but who knows?
is your documentation original? perhaps you can snip out a larger portion of the drawing showing what your questions are about...
in the course of manufacturing a product, several engineering revisions are not uncommon and is based on feedback coming from the field or perhaps better materials and parts that they wanted to upgrade with the revisions....
so i am not surprised to see on boards many blank holes that parts were not stuffed in..
in the past, i would email the manufacturer with my technical queries stating the model number and serial number of the unit, sometimes i get lucky and got a reply...
blocking caps at the output of the op-amps are done to eliminate the possible damage it may cause to the power amp it feeds if the power amp is direct coupled at the input..
in most cases may not be needed, but who knows?
is your documentation original? perhaps you can snip out a larger portion of the drawing showing what your questions are about...
in the course of manufacturing a product, several engineering revisions are not uncommon and is based on feedback coming from the field or perhaps better materials and parts that they wanted to upgrade with the revisions....
so i am not surprised to see on boards many blank holes that parts were not stuffed in..
Yeah, that schematic when printed or even on my monitor is useless. I couldn't make out anything!
I found that viewing the pdf on my phone (Galaxy Note 8) I was able to zoom in with much, much greater resolution and was able to read the schematic. Though, only a couple inches at a time 😉
- Status
- Not open for further replies.