I'm still following all this 🙂 Thanks for clarifying the readings earlier.
Can the op tell us whether this amp failed in service or has it been an acquisition with an unknown history? If it is the latter then could any parts have been replaced with incorrect values? Although very unlikely, resistors can occasionally go low in value and R102 dropping in value would possibly give the symptoms reported.
The midpoint adjustment preset I assume is on minimum resistance which translates to trying to increase the mid point voltage to its maximum value.
If TR100 is removed the voltage on what was the emitter (so C101 voltage) should be about 10 volts and is a good sanity check on the biasing of the front end. This also proves R104 and R102 are delivering the correct biasing voltage to the emitter.
The midpoint would fall to a couple of volts in this scenario and crucially the voltage on what would be the base of TR100 should be around 3 volts and should be identical on either side of R108 (the 22k feedback resistor). This is another check that shows nothing is pulling one side down.
The voltage on R111 should have fallen to about 3 to 3.5 volts and will be a little lower on one side than the other (preset side will be the lower)
Can the op tell us whether this amp failed in service or has it been an acquisition with an unknown history? If it is the latter then could any parts have been replaced with incorrect values? Although very unlikely, resistors can occasionally go low in value and R102 dropping in value would possibly give the symptoms reported.
The midpoint adjustment preset I assume is on minimum resistance which translates to trying to increase the mid point voltage to its maximum value.
If TR100 is removed the voltage on what was the emitter (so C101 voltage) should be about 10 volts and is a good sanity check on the biasing of the front end. This also proves R104 and R102 are delivering the correct biasing voltage to the emitter.
The midpoint would fall to a couple of volts in this scenario and crucially the voltage on what would be the base of TR100 should be around 3 volts and should be identical on either side of R108 (the 22k feedback resistor). This is another check that shows nothing is pulling one side down.
The voltage on R111 should have fallen to about 3 to 3.5 volts and will be a little lower on one side than the other (preset side will be the lower)
Well it needs to be stated that the paragraph concerned by JLH is complete nonsense. He utters contradictions in terms such as crossover distortion in Class A, and doesn’t seem to realise that class A is a whole 180 degrees of conduction angle away from class B. You can’t get a 303 into class A at all. Never intended to get anywhere near it. You would need about ten times the heat sinking.
It needs to be stated that JLH was talking about quiescent current - that is the level of operation at the crossover point of the output stage. The point of a class AB amplifier -is to bias the output so the level of output stage current is sufficient to bridge the crossover region. In this case the Class A power level could be measured in milli-watts.
The transconduction plot of transistors is curved between 0.53 and 0.6 volts Vbe. Beyond 0.6 volts this increasingly takes the form of a steep upward slope .i.e. linear as opposed to a curve i.e. non-linear.
The transconduction plot of transistors is curved between 0.53 and 0.6 volts Vbe. Beyond 0.6 volts this increasingly takes the form of a steep upward slope .i.e. linear as opposed to a curve i.e. non-linear.
Thanks guys, currently 2000 miles away but when back will pick this ip again
Re post 17, you may have missed the point of measuring the original R130 out of circuit.
The value of R130 is 150k it should not fail through burnout or go high in value due to overheating in the circumstances here.
The possiblities exist of a discontinuity somewhere in series with the adjustment setting components such as a crack in the traces under the pcb or a cold solder joint.
Last edited:
There are times a high value resistor simply increases in value, or the end caps fail. But you are correct, you can only be sure you have the right measurement with the component isolated at least with only one lead connected to circuit.
No it isn’t. It is the current drawn when there is no signal.
In the case of the Quad 303 there is no Class A power level whatsoever.
You’re just adding to the confusion. Obviously JLH should have written AB where he wrote A, but that’s what he wrote, and published, and it doesn’t enhance his credibility.
One could try measuring the total resistance (continuity differences) from the top of R130 through intermediate resistors in the adjustment line to earth and do the same for the other channel.
Last edited:
Absolutely no signs of visual stress on this resistor, good point on a crack and will buzz the board out to be sure
The amp was an acquisition unknown condition, when visually inspecting the boards components matched perfectly.
R.e the above, I will try this tomorrow.
Measured voltages r.e above post today, they are
Emitter of Tr100 - 7.6vdc
R108 / R111 3.3vdc both sides
R101 around 3.3vdc both sides
Center voltage 1.6vdc
Emitter of Tr100 - 7.6vdc
R108 / R111 3.3vdc both sides
R101 around 3.3vdc both sides
Center voltage 1.6vdc
Nevertheless they do go high, as do many high value resistors connected to HT. It’s a well-known phenomenon.
That looks low. So this is with TR100 removed and we have a potential divider of 22k and 3.9k and a supply of 67 volts. That should give about 10 volts on the emitter pad.
22k+3k9 = 25.9k 67 volts across that generates 10 volts across the 3k9. That needs investigating first. Check the values fitted and check the supply onto the 22k.
We'll take that as OK and same for R101
Investigate TR100 emitter volts.
I cannot see any series feedback loop and there appears to be one for shunt feedback for dc purposes if this loop is connected to the midpoint of the output triples. This is why I made a comment about investigating these structures. I have not been able to find a full version of the circuit..
I have seen preamp schemes elsewhere where an output is connected to ground by two equal resistor values and the midpoint between these resistors is taken from there via a third resistor to the input transistor base. If there is some anomalous fault in the triplets, that could have dominant effects for preceding bias set up stages.
According to oneQuad authority the 303 inverts the input signal.
I have seen preamp schemes elsewhere where an output is connected to ground by two equal resistor values and the midpoint between these resistors is taken from there via a third resistor to the input transistor base. If there is some anomalous fault in the triplets, that could have dominant effects for preceding bias set up stages.
According to oneQuad authority the 303 inverts the input signal.
Last edited:
There isn’t one.
It is for both AC and DC purposes.
The Quad 303 circuit is all over the Internet. Keep looking.
Hard to see what that could be, or what a balanced output from a preamp has to do with it.
You don’t need an authority. It is. Just count the common-emitter stages.
@ejpthanks for confirming the point about the absence of a series feedback loop. That may not be obvious to new members.
Re shunt feedback while that is for ac and dc purposes the former does not count when there is a dc issue to be sorted. So far in present investigations it seems that no fault has been remedied outside of the output triplets, the advice given to me about the schematic of keep on looking is not turning up trumps.
I might have expected some questions about the fact the 303 inverts the input signal and this might have been something I had dreamed up. You raised the point about counting common emitter amplifiers, but you stopped short of identifying these which is less than objective. You seem to be posing a test of knowledge as a trap for me. I am calling your bluff mate.What are these.
Re shunt feedback while that is for ac and dc purposes the former does not count when there is a dc issue to be sorted. So far in present investigations it seems that no fault has been remedied outside of the output triplets, the advice given to me about the schematic of keep on looking is not turning up trumps.
I might have expected some questions about the fact the 303 inverts the input signal and this might have been something I had dreamed up. You raised the point about counting common emitter amplifiers, but you stopped short of identifying these which is less than objective. You seem to be posing a test of knowledge as a trap for me. I am calling your bluff mate.What are these.
I noted that some of the measured voltage measurements between the good and bad channel are roughly equivalent to two diode voltage drops ,and that John Ellis posted a revised circuit in 2007 where he omitted two of these, and substituted a Vbe multiplier stage. He also used fresh part number designations in the process making comparisons less easy.
I don't know if this provides any clues as to whether John's revisions and if these are related to the present situation.
As far as common emitter amplifiers go there is a positive feedback path from the bootstrap load to the Vas transistor to the base of the BC154 so this is now in series and in normal phase with the input and this should boost the input and output from this device -so take due care not to overload the base of this transistor. If one of the bias diodes is faulty that could lead to other problems.
There is two diode biasing for TR102 here. There is also diode biasing for the output stages in combination with short circuit diode protection. So far these areas have not been compared between the good a bad channel.
I don't know if this provides any clues as to whether John's revisions and if these are related to the present situation.
As far as common emitter amplifiers go there is a positive feedback path from the bootstrap load to the Vas transistor to the base of the BC154 so this is now in series and in normal phase with the input and this should boost the input and output from this device -so take due care not to overload the base of this transistor. If one of the bias diodes is faulty that could lead to other problems.
There is two diode biasing for TR102 here. There is also diode biasing for the output stages in combination with short circuit diode protection. So far these areas have not been compared between the good a bad channel.