I need some help with my new (to me) HP 8903B distortion analyzer. Unit works great and is spot on for all of the tests in section 4 of the manual.
I checked a friend's tube amp with it to measure distortion about two months ago and it worked great. The tube amp was a three wire power cord, so I'm assuming the chassis was earth grounded, but not sure if the audio circuit was referenced to earth ground.
I finally got new equipment shelves assembled and last night I wanted to check one of my Zen amplifiers I built several years ago. Something is not correct and it looks like a ground loop maybe? The Zen amp does not use an earth ground because it is in a wooden case and the circuit does not need an earth reference. I feed the signal output of the 8903 to the input of the amp (using the high output in non-float mode) to the single ended input of the Zen amp. I then have an 8 ohm load connected to the amp output and feed that loaded output to the 8903 high input (non-float mode). The distortion is high at very low levels of signal and the monitor output shows the bottom half of the sine wave being squashed. I put a speaker on the output instead of a load and noticed when I remove the connection to the input of the 8903 the sound gets louder and the audible distortion diminishes. I tried just looking at the output of the amp directly with the scope and noticed that when I touch the BNC connector shield to the scope, it distorts the sound from the speaker and drops the level. Also, the sine wave compresses. The same occurs with the when connecting to the input of the 8903.
I used to use a leader function generator to generate sine waves, so I pulled this out and connected it to the input of the amp and looked at the sine wave on the scope. The sine wave was fine and the signal did not change when the scope was connected / disconnected. The Leader generator is only a two wire device. I'm thinking I have a ground loop??? Everything is plugged into the same outlet on one circuit.
It became very late last night, so I did not pull out another amp to validate my findings. Any help would be appreciated.
Thanks! Dave
I checked a friend's tube amp with it to measure distortion about two months ago and it worked great. The tube amp was a three wire power cord, so I'm assuming the chassis was earth grounded, but not sure if the audio circuit was referenced to earth ground.
I finally got new equipment shelves assembled and last night I wanted to check one of my Zen amplifiers I built several years ago. Something is not correct and it looks like a ground loop maybe? The Zen amp does not use an earth ground because it is in a wooden case and the circuit does not need an earth reference. I feed the signal output of the 8903 to the input of the amp (using the high output in non-float mode) to the single ended input of the Zen amp. I then have an 8 ohm load connected to the amp output and feed that loaded output to the 8903 high input (non-float mode). The distortion is high at very low levels of signal and the monitor output shows the bottom half of the sine wave being squashed. I put a speaker on the output instead of a load and noticed when I remove the connection to the input of the 8903 the sound gets louder and the audible distortion diminishes. I tried just looking at the output of the amp directly with the scope and noticed that when I touch the BNC connector shield to the scope, it distorts the sound from the speaker and drops the level. Also, the sine wave compresses. The same occurs with the when connecting to the input of the 8903.
I used to use a leader function generator to generate sine waves, so I pulled this out and connected it to the input of the amp and looked at the sine wave on the scope. The sine wave was fine and the signal did not change when the scope was connected / disconnected. The Leader generator is only a two wire device. I'm thinking I have a ground loop??? Everything is plugged into the same outlet on one circuit.
It became very late last night, so I did not pull out another amp to validate my findings. Any help would be appreciated.
Thanks! Dave
Measuring floating output amps can be a bit tricky but it sounds you have everything you need. The output of the generator can be set to floating or grounded - see what works best.
The outputs of the amp goes to the center pin of each bnc input connector on the analyzer. So you will use both bnc inputs to measure your zen amp with the floating output.
The analyzers input should be set to floating.
Wasn't very clear from you question above if this is what you had tried and if so, maybe a diagram of the wiring and the amp in question will help. Lots of 8903b users here to lend a hand. G luck!
The outputs of the amp goes to the center pin of each bnc input connector on the analyzer. So you will use both bnc inputs to measure your zen amp with the floating output.
The analyzers input should be set to floating.
Wasn't very clear from you question above if this is what you had tried and if so, maybe a diagram of the wiring and the amp in question will help. Lots of 8903b users here to lend a hand. G luck!
When you use the BNC connections on the 8903B, ONLY the center pin is intended to be connected to the device under test (DUT). The ground of the BNC is there for cable shielding, and is not to be connected to the DUT.
The 8903B originally included a set of BNC to single binding post adapters that look a lot like the ones in the attached pic (I couldn't find a picture of the original). These simplified connections when not using shielded cables, and they expose only the center conductor of the BNC; the shield does not come out to another binding post because as I said, it is not intended to be connected to the DUT.
To make the input connections properly, you can use adapters like the picutred ones, or make a set of cables by cutting a BNC cable in half, then attaching a banana plug to the center conductor of each (don't connect the shield to anything). Float the input of the 8903B; there is almost never a situation when you would not float it, especially when testing an amp rather than line-level gear.
For the 8903B output, you can set it to grounded, which grounds the LOW output center pin, forcing the differential output to act as single ended. Then you can use the HI side BNC to connect in the usual manner to the RCA input jack of the amp.
The 8903B originally included a set of BNC to single binding post adapters that look a lot like the ones in the attached pic (I couldn't find a picture of the original). These simplified connections when not using shielded cables, and they expose only the center conductor of the BNC; the shield does not come out to another binding post because as I said, it is not intended to be connected to the DUT.
To make the input connections properly, you can use adapters like the picutred ones, or make a set of cables by cutting a BNC cable in half, then attaching a banana plug to the center conductor of each (don't connect the shield to anything). Float the input of the 8903B; there is almost never a situation when you would not float it, especially when testing an amp rather than line-level gear.
For the 8903B output, you can set it to grounded, which grounds the LOW output center pin, forcing the differential output to act as single ended. Then you can use the HI side BNC to connect in the usual manner to the RCA input jack of the amp.
Attachments
Guys, thanks for the feedback. I do have the BNC to banana adapters that originally came with the 8903. I'll try tonight and see what I can resolve. It got to be too late last night and I was too tired.
I'll post back with results.
I'll post back with results.
Success. I floated the output and there were no issues measuring the amp. I did leave the input referenced to ground and used the high output. I also tested another amp (manufacturer produced) this way with good success. Thanks for help!
Dave
Dave
- Status
- Not open for further replies.