I am getting a strange distortion from both channels of my pre-amp (Harman Kardon Signature 2.0). I had not been using it since more than a year now. When I recently started using it, I started getting a strange kind of distortion, as if the speaker cone is torn. The vocals sound torn especially with recordings of a high input level. I tried eliminating every piece of equipment & interconnect in the line-up & the problem lies with the preamp. Now it's getting worse, with the balance shifting a lot towards the left channel. The left channel sounds louder & brighter than the right. Please Suggest 
According to page 14 of the service manual of that amp, the volume control is electronic, i.e. the knob doesn't operate a conventional potentiometer.
I take it you tried every input on the preamp and get the same result on each input?
Since input selection and volume control seem all electronic in this device, I'd start with measuring if all voltages in the PSU are present and correct.
I take it you tried every input on the preamp and get the same result on each input?
Since input selection and volume control seem all electronic in this device, I'd start with measuring if all voltages in the PSU are present and correct.
I just peeked at the service manual as well and you'll have quite the number of components to check! Like jitter told, using the manual, knowing a bit about the used ICs, you could check up on all the voltages to see if they are proper; measure the supply output and if that is ok, measure the supply pins of various ICs throughout and if that looks okay, try to find any capacitors that may look like they're puffed up or leaking or otherwise show anomalies.
It's going to be quite the hunt.
Edit: Oh and forgot to mention another 'easy' checkup: Internal connectors gone bad (bad contact) so reseating those may also help.
It's going to be quite the hunt.
Edit: Oh and forgot to mention another 'easy' checkup: Internal connectors gone bad (bad contact) so reseating those may also help.
Last edited:
Your explanation was spot-on. Short & clear. That being said, since I do not have a test-tome generator, can I use my computer in which I have a software-based test tone generator to connect it to the input of my preamp & check the way you mentioned? What setting does the Scope need to be at??
The test-tone generator software generates Sine waves Square waves, pink noise, white noise etc. I guess it's the same as using a Hardware tone generator isn't it??
The test-tone generator software generates Sine waves Square waves, pink noise, white noise etc. I guess it's the same as using a Hardware tone generator isn't it??
Yes you can.
I would suggest you start by reading a tutorial on how to use a scope, e.g. like this one. This particular tutorial is based on a Hameg analogue oscilloscope, but are applicable to all scopes. Apart from different looks and position of buttons and knobs, they work the same. A digital scope usually has an "autoset" button, if there's a good signal on the input a single push will usually be all that's needed to set the scope.
Try it out first on the (analogue) output of the soundcard and set the software for a sinewave of 1 kHz, 1 Vpp, see if you can get it on the screen of the scope.
At 0.5 V/div, the sinewave should be across two vertical divisions of the screen. If you don't get this, see if the volume control in the operating system of the PC isn't set too low.
Also be aware that a 1:10 probe (should be printed on the probe) attenuates the signal to 1/10 th, and all readings on the scope should be multiplied by 10 to get the correct value. In a 1:1 probe, there's no attenuation.
A single period of a 1 kHz (1/freq) is 1 ms, so if you set the horizontal scale for 0.2 ms/div, you should view one full period across 5 divisions.
If you feel comfortable using the scope, start tracing as suggested. Keep in mind that in the amp the signal might get attenuated or amplified by the several circuits, so you may need to turn the Volts/div knob quite a bit to keep a good view. Also keep in mind that very low level signals in the lowest ranges of the scope (5 mV or 10 mV/div) might look ragged as a result of noise "polluting" the measurement, don't automatically assume it's the amp doing that.
Another thing: there will probably be lots of circuits with DC on them, so keep the input switched to AC for now (DC/AC/GND switch close to input).
I would suggest you start by reading a tutorial on how to use a scope, e.g. like this one. This particular tutorial is based on a Hameg analogue oscilloscope, but are applicable to all scopes. Apart from different looks and position of buttons and knobs, they work the same. A digital scope usually has an "autoset" button, if there's a good signal on the input a single push will usually be all that's needed to set the scope.
Try it out first on the (analogue) output of the soundcard and set the software for a sinewave of 1 kHz, 1 Vpp, see if you can get it on the screen of the scope.
At 0.5 V/div, the sinewave should be across two vertical divisions of the screen. If you don't get this, see if the volume control in the operating system of the PC isn't set too low.
Also be aware that a 1:10 probe (should be printed on the probe) attenuates the signal to 1/10 th, and all readings on the scope should be multiplied by 10 to get the correct value. In a 1:1 probe, there's no attenuation.
A single period of a 1 kHz (1/freq) is 1 ms, so if you set the horizontal scale for 0.2 ms/div, you should view one full period across 5 divisions.
If you feel comfortable using the scope, start tracing as suggested. Keep in mind that in the amp the signal might get attenuated or amplified by the several circuits, so you may need to turn the Volts/div knob quite a bit to keep a good view. Also keep in mind that very low level signals in the lowest ranges of the scope (5 mV or 10 mV/div) might look ragged as a result of noise "polluting" the measurement, don't automatically assume it's the amp doing that.
Another thing: there will probably be lots of circuits with DC on them, so keep the input switched to AC for now (DC/AC/GND switch close to input).
Last edited:
You'll need to attach it to a ground connection close to where you want to measure, you could use the grounded leads of components on the board; the connections that connect to the ground plane / ground tracks on the PCB.
As for setting it to DC/AC, that matters on whether you are interested in measuring the DC component of the signal. Since you measure against the ground, the DC component can make the wave go out of view on the scope. You can turn a knob to get it back into view but then the reading won't make sense voltage wise. So then, you set it to AC to filter out the DC component and have a signal center on the Y axis. It's basically switching a capacitor in series with your probe.
As for setting it to DC/AC, that matters on whether you are interested in measuring the DC component of the signal. Since you measure against the ground, the DC component can make the wave go out of view on the scope. You can turn a knob to get it back into view but then the reading won't make sense voltage wise. So then, you set it to AC to filter out the DC component and have a signal center on the Y axis. It's basically switching a capacitor in series with your probe.
In addition to MagicBox's remarks:
This is where reading the safety instructions becomes important.
If a circuit common floats at a highish voltage above earth potential, then connecting the black lead to that circuit common will be dangerous. Measurements on these kinds of circuits must be taken with a differential probe.
Having said that, above case is unlikely in an amp. It will have circuit common connected to the chassis, so it will be ~0 V. It's best to clip the test lead onto the chassis. For ease of operation, you might instead run a separate lead from amp chassis to ground on the scope (outside of a BNC-connector or the testpin marked with an earth symbol, to the left of channel 1 input). In that case, unclip the black testlead from the probe so that it cannot dangle and inadvertantly cause a short.
This is where reading the safety instructions becomes important.
If a circuit common floats at a highish voltage above earth potential, then connecting the black lead to that circuit common will be dangerous. Measurements on these kinds of circuits must be taken with a differential probe.
Having said that, above case is unlikely in an amp. It will have circuit common connected to the chassis, so it will be ~0 V. It's best to clip the test lead onto the chassis. For ease of operation, you might instead run a separate lead from amp chassis to ground on the scope (outside of a BNC-connector or the testpin marked with an earth symbol, to the left of channel 1 input). In that case, unclip the black testlead from the probe so that it cannot dangle and inadvertantly cause a short.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.