How do you compare complementary transistors these days; are you able to do the comparison simultaneously or you have to do it individually, print data and do next? Although past discussion focused on comparing transistors of similar (substitute) on curve tracers using such features as A / B (Left / Right) selector switch on Tektronix and Leader like instruments, I am interested finding out if it is possible to validate complementary pairs simultaneously. In this case we are looking at comparing a PNP vs NPN for characteristic comparison simultaneously.
Run them in a complimentary amp.
Some of my designs I have to match the genders , to match the P/N is nearly impossible , you see
how far off you are by offset in the completed amp.
I can match a P/N to say 300hfe , but the design will still be off. I just make the design less dependent on any imbalance.
A way around this is to use a dual monolithic matched (usually SMD) device.
Even with a pre-matched pairs , the circuit is still "off" as P/N (same Hfe) always has different V/I characteristics.
With outputs , the genders are always different.
OS
Some of my designs I have to match the genders , to match the P/N is nearly impossible , you see
how far off you are by offset in the completed amp.
I can match a P/N to say 300hfe , but the design will still be off. I just make the design less dependent on any imbalance.
A way around this is to use a dual monolithic matched (usually SMD) device.
Even with a pre-matched pairs , the circuit is still "off" as P/N (same Hfe) always has different V/I characteristics.
With outputs , the genders are always different.
OS
ostripper thanks for response.
Although I started out in custom designing of pro-audio for musicians (Italy), later on it evolved into a full-blown service center representing 23 US, British, German, Norway, New Zealand and Japanese consumer and professional audio and video. I become among the few Yamaha Preferred Service Centers and major Onkyo, Denon, B&O, Carver, Pioneer, ADCOM, Marantz, GAS, Carver etc. service center. For professional AV I became Panasonic Pro-Video (including their early 80's cellphone) and JVC service center.
Over time having made close relationship with semiconductor manufacturers like Panasonic, Toshiba and Mitsubishi (my favorite) and taking advice from their respective field engineering representatives the following setup was considered and applied which I still have.
From your statement ... run them in a complimentary amp; also implies that one could run two curve tracers like Tektronix 7CT1N simultaneously plugged into one 7603 but send the curve tracers X and Y outputs to two separate external OSD oscilloscopes. One 7CT1N will test a PNP and while it is in place the second 7CT1N can be used to select a transistor that is closely matching as pair. There are two curves worth paying attention when comparing two complementary transistors.
The first Step in the curve closer to zero should be similar both as much as possible. This comparison is to make sure both transistors respond similarly when base voltage is low. This is very important for drivers as well since their base voltage often is trimmed by bias network. The last desired 4th or 5th step (in my case) is compared to see if the steps (amplitude) progressively get larger from one another with increasing number of steps. The variation in the steps can be approximated very quickly between the transistors using the delta-V cursor measurement.
The above test is comparing the complementary state of 2SB755-R vs 2SD845-R. Same setup is being used also for differential transistors, & FETs, drivers and just about any would be complementary or differential. A similar setup also compares small diodes.
Although I started out in custom designing of pro-audio for musicians (Italy), later on it evolved into a full-blown service center representing 23 US, British, German, Norway, New Zealand and Japanese consumer and professional audio and video. I become among the few Yamaha Preferred Service Centers and major Onkyo, Denon, B&O, Carver, Pioneer, ADCOM, Marantz, GAS, Carver etc. service center. For professional AV I became Panasonic Pro-Video (including their early 80's cellphone) and JVC service center.
Over time having made close relationship with semiconductor manufacturers like Panasonic, Toshiba and Mitsubishi (my favorite) and taking advice from their respective field engineering representatives the following setup was considered and applied which I still have.
From your statement ... run them in a complimentary amp; also implies that one could run two curve tracers like Tektronix 7CT1N simultaneously plugged into one 7603 but send the curve tracers X and Y outputs to two separate external OSD oscilloscopes. One 7CT1N will test a PNP and while it is in place the second 7CT1N can be used to select a transistor that is closely matching as pair. There are two curves worth paying attention when comparing two complementary transistors.
The first Step in the curve closer to zero should be similar both as much as possible. This comparison is to make sure both transistors respond similarly when base voltage is low. This is very important for drivers as well since their base voltage often is trimmed by bias network. The last desired 4th or 5th step (in my case) is compared to see if the steps (amplitude) progressively get larger from one another with increasing number of steps. The variation in the steps can be approximated very quickly between the transistors using the delta-V cursor measurement.
The above test is comparing the complementary state of 2SB755-R vs 2SD845-R. Same setup is being used also for differential transistors, & FETs, drivers and just about any would be complementary or differential. A similar setup also compares small diodes.
When do you need to worry about a transistor characteristic behaviour near the low end of ramp test using Tektronix or similar test instrument? Allow me to use one of my favorite Technics amplifier SU-V8.
Like all electronics service, you start by checking the power supply voltage to make sure they are approximately equal. If the problem is serious and you suspect a possible loss of expensive outputs, this particular amplifier allows the testing of the voltage amplifier stage without the outputs in place. It is good to identify the drivers and their associated voltages with no signal going through. Remember the third image shows outputs removed and what you may have to do to stabiles the references (red resistors) feed back to Q602 and Q604. In third figure the final drivers emitter DC voltage should be about 0.6 Vdc.
Where does the problem of hFE test help in components. Transistors in IDLE state operate at lowest switched state. While most diodes in this amplifier are fairly stable, the first place to verify if the DC voltages are correct will be in the Pre-Driver PCB (first image). The connectors label L and N should not exceed 2.0 and -2.0 volts, assuming connector M (feedback) is zero. Q316 and Q314 need to be tested at higher voltage and at least 5 to 7 step function in Tektronix & they should have a similar behaviour near 60 Vdc. After that all transistors in the driver and output stages all need a closer look at the transistors behaviour at first STEP. Every other STEP interval can be compared using the cursor on your DSO.
Some technicians not familiar with this Amplifier and as well as many others Like Onkyo TX-NR900 may be that drivers are getting hot. Don't go by the temperature in this case, monitor carefully the idle state voltage to make sure they are stable. Turn the volume UP quarter way and back down quickly to see if the BIAS remains stable. If the base voltages to the outputs remain constant and balanced (offset wise), then it is time to test the amplifier THD. THD test is done for short duration at or near peak power to avoid damage.
Where does the problem of hFE test help in components. Transistors in IDLE state operate at lowest switched state. While most diodes in this amplifier are fairly stable, the first place to verify if the DC voltages are correct will be in the Pre-Driver PCB (first image). The connectors label L and N should not exceed 2.0 and -2.0 volts, assuming connector M (feedback) is zero. Q316 and Q314 need to be tested at higher voltage and at least 5 to 7 step function in Tektronix & they should have a similar behaviour near 60 Vdc. After that all transistors in the driver and output stages all need a closer look at the transistors behaviour at first STEP. Every other STEP interval can be compared using the cursor on your DSO.
Some technicians not familiar with this Amplifier and as well as many others Like Onkyo TX-NR900 may be that drivers are getting hot. Don't go by the temperature in this case, monitor carefully the idle state voltage to make sure they are stable. Turn the volume UP quarter way and back down quickly to see if the BIAS remains stable. If the base voltages to the outputs remain constant and balanced (offset wise), then it is time to test the amplifier THD. THD test is done for short duration at or near peak power to avoid damage.
