Perry if I find a great deal on a digital scope should I get it. What’s your reason for
preferring analog scopes?
preferring analog scopes?
The waveform display on an analog scope is far superior to that of an LCD display. I just can't stand the donkey kong display of the LCD. There are some LCD scopes that get close but will never be as good until they increase the resolution and data recorded. The tech is there for the display. For the data, it may be too costly. I had a digital display scope but didn't use it very often. I had some digital scopes that worked in both modes and those were nice but they didn't get used much either. My favorite scope was a Tek 465B with a DM44 option (the display rarely got used). For me, that scope was absolute perfection but one in good condition (complete refurb and calibrated) is 3x the cost of an entry level digital scope... and for me, would be worth the cost. For someone who just needed a basic scope, the price couldn't be justified.
Thanks Perry. I'll see the condition of the analog scopes and report back. Do you think it's a worthwhile purchase for this repair project or would you recommend the amp driver swap first?
If you do this type of work without a scope, you're essentially working blind. You may get lucky and get everything right without one but there's a reason every professional tech owns (at least) one.
Swapping the board costs nothing more than solder and braid.
At this point, I can't even suggest anything to check with a scope.
Swapping the board costs nothing more than solder and braid.
At this point, I can't even suggest anything to check with a scope.
Hi Perry and all. Have a few questions on setting up an affordable but useful test bench. What are your thoughts on the Hitachi V422 oscilloscope? It seems very basic and I feel I will be doing lots of manual calculations and adjustments but it's <$100. Also, I've been reading that I should use a signal generator. Can I just use my phone with a signal generator app connected with an adapter to the amps RCAs? Also, do I need a dummy load like an 8-ohm 100W resistor? If so, does it have to be non-inductive? Also in my research, I came across a safety recommendation to use an isolated transformer. Currently, I am using a PC power supply in which the ground going to the amp is grounded to earth.
One more question. I see tons of ebay oscilloscopes that say they power on but no other testing was done. How safe is it to buy these? How common is it for oscilloscopes to have issues but power on fine?
Do NOT buy a scope that doesn't have a locked waveform displayed on the display unless it has a way to get all money back (including shipping).
What calculations?
You can use your phone but I'd suggest grounding the shields to the PS ground via a 12v lamp (preferably an 1156 or 1157 lamp).
The 422 will be perfectly fine if in good working order.
You will need a way to test the amp. If you want to do it on the bench, you will need a speaker, dummy load or both.
It doesn't matter if it's inductive or not. The inductance will be minimal.
Don't use an isolation transformer. They are used by TV techs working on old TVs that have a live chassis.
What calculations?
You can use your phone but I'd suggest grounding the shields to the PS ground via a 12v lamp (preferably an 1156 or 1157 lamp).
The 422 will be perfectly fine if in good working order.
You will need a way to test the amp. If you want to do it on the bench, you will need a speaker, dummy load or both.
It doesn't matter if it's inductive or not. The inductance will be minimal.
Don't use an isolation transformer. They are used by TV techs working on old TVs that have a live chassis.
Sorry what do you mean by a locked waveform? Some I have only seen a flat line shown in the listing since they have not tested it with probes on any signal. Just turned it on.
From my little knowledge of oscilloscopes, it seems these older ones like the 422 you have to calculate the voltage for the y axis and the seconds (frequency) on the x axis for each test point you perform. On newer oscilloscopes, they do the calculations automatically and center the waveform, and give you information about the waveform on the screen.
Also not sure I understand the grounding of the RCA shields with a lamp.
From my little knowledge of oscilloscopes, it seems these older ones like the 422 you have to calculate the voltage for the y axis and the seconds (frequency) on the x axis for each test point you perform. On newer oscilloscopes, they do the calculations automatically and center the waveform, and give you information about the waveform on the screen.
Also not sure I understand the grounding of the RCA shields with a lamp.
Last edited:
There is no reason not to have a square wave clearly displayed on the scope since most all scopes have a test signal used to calibrate probes.
For audio, you use 2ms. For PS waveforms, you typically use 10us and have 3 cycles on the display. For the class D amps, you use 5us and have 3-5 cycles on the display.
You don't use a scope to measure voltage accurately. For waveforms, you know the approximate voltage you should have and can see it on the scope's display. FETs typically use a 10v amplitude square wave drive. For audio, you just need to see that it's clean. You only RARELY need to make any calculations. For frequency, you can use a multimeter if you need to know a precise frequency (rare).
You don't absolutely need to ground the shield but it's better if you do.
For audio, you use 2ms. For PS waveforms, you typically use 10us and have 3 cycles on the display. For the class D amps, you use 5us and have 3-5 cycles on the display.
You don't use a scope to measure voltage accurately. For waveforms, you know the approximate voltage you should have and can see it on the scope's display. FETs typically use a 10v amplitude square wave drive. For audio, you just need to see that it's clean. You only RARELY need to make any calculations. For frequency, you can use a multimeter if you need to know a precise frequency (rare).
You don't absolutely need to ground the shield but it's better if you do.
Havent been able to find an oscilloscope yet. I tried connecting the 100uf 100v bipolar capacitor in line with either the positive or negative speaker terminal and the 32vdc still go through. That one transistor on the amp driver board heats up super fast. Within seconds it’s at 120F! Once I have the oscilloscope how will I be able to test the amp with this transistor heating up? What’s the plan for testing with the oscilloscope? I guess we will be sending a signal through the RCA inputs of the amp and then test at different places if the signal is correct? Where do I start from? We already know the speaker terminals have the rail voltage. Do I follow the schematic back from there? How do I know which spots to choose to test from? Also where does the ground go? To audio ground, chassis or earth ground? Also would it be possible to feed the calibration square wave that most oscilloscopes have to the amp? Kinda like using it as a signal generator? They are usually 1khz which is a good frequency to use.
Been reading up and watching tons of videos on using oscilloscopes including Perry’s own page on the topic. For others that want to learn I highly recommend these:
https://www.bcae1.com/oscope.htm
https://youtu.be/zhON9mAL2KA
https://youtu.be/7T8hxhpHua4
https://youtu.be/KkJ_WOVWQWM
https://youtu.be/f6GUM71Y9aI
Been reading up and watching tons of videos on using oscilloscopes including Perry’s own page on the topic. For others that want to learn I highly recommend these:
https://www.bcae1.com/oscope.htm
https://youtu.be/7T8hxhpHua4
https://youtu.be/KkJ_WOVWQWM
https://youtu.be/f6GUM71Y9aI
Sorry if I misunderstood…So your recommendation is to swap the drive boards first? Or should I continue and remove Q2-Q6 to test them outside the circuit?
I think that's the first step (to swap the driver boards).
The fact that there is an intermittent problem with only one output transistor is strange.
The amp may have multiple problems.
The fact that there is an intermittent problem with only one output transistor is strange.
The amp may have multiple problems.
Removing the driver boards was extremely easy thanks to your method Perry. Much easier than removing the output transistors. I swapped the boards and the problem didn’t move but instead stayed on the right channel. In order to test I soldered the broken leg of the output transistor.
With the transistor that you (at one time) thought was bad, out of the circuit, what does the amp do that it should not do?
Removed Q2. The base and collector pins of Q2-Q6 were not shorted anymore. Immediately after powerup there was no dc voltage on the right channel. After about 15-20 seconds the 32Vdc started. I shut down the amp and now the base and collector pins on Q2-Q6 are shorted again.