I am attempting to use a non-inverting LM3875 amp with Speaker Workshop from my laptop to do some testing. I intentionally attenuated the signal at the amp's input using a 4:1 voltage divider, then to a 100k pot also configured as a voltage divider. I did this to increase the sensitivity of the pot within a small range and to keep the chip cool, as I used a small heatsink since I wanted the whole rig to be easily transportable.
The whole thing works great when a regular CD player is my source. I get around 4 watts into an 89-dB speaker (using the RS sound level meter) and the chip stays cool enough to run for hours without going into thermal protection. Today, I hooked it up to my laptop and using the line out went to calibrate it using the Eric Wallin method, specifically using a DMM to obtain 2.83V into 8 ohms resulting in 1W at 1000Hz. I never could get above .98V, however.
After double checking many things, I tried to check the voltage at the output of the line out using the DMM and the 1kHz signal. On the laptop, it was no more than .068V. I checked the output on my CD player and it was 2.5V, using a 1kHz test tone from a test CD.
My questions are:
1) Is this the correct method for determining the source output voltage, or am I presenting a load with the DMM that is too much for the laptop to handle. I assume the impedance of the DMM would be on the 100 or so kHz level, so I doubt this would cause any problem.
2) I have always heard that source output voltages are typically assumed to be around 1.5V when designing the gain for a generic DIY amplifier. Am I mistaken in this assumption? If that is a typical assumption, then should I be surprised that my CD player was outputting 2.5V? I have no reference as to the recording levels on the CD, so this could be a variable that is not correctly accounted for.
3) Since the gain of the LM3875 should typically be above 10, from what I've read here, which seems to be a result of the required Gain Bandwidth Product from the National Data Sheet, is the method of using a voltage divider ahead of the volume control attenuator a good practice? At this time, I used the RJM schematic for the non-inverting amp, so the actual gain is closer to 32. I could cut this down and reduce the drop across the voltage divider, but I think it would still be required since I am attempting to keep the output of the amp low. Is there a better suggestion for accomplishing this goal?
Thanks for reading through this long post. I thought the chip amp forum would be the most appropriate place to post, even though it is a multi-faceted question regarding source components, speaker measurement and the chip amp as well.
Thanks in advance for any help which you can provide.
Best Regards,
Sandy.
The whole thing works great when a regular CD player is my source. I get around 4 watts into an 89-dB speaker (using the RS sound level meter) and the chip stays cool enough to run for hours without going into thermal protection. Today, I hooked it up to my laptop and using the line out went to calibrate it using the Eric Wallin method, specifically using a DMM to obtain 2.83V into 8 ohms resulting in 1W at 1000Hz. I never could get above .98V, however.
After double checking many things, I tried to check the voltage at the output of the line out using the DMM and the 1kHz signal. On the laptop, it was no more than .068V. I checked the output on my CD player and it was 2.5V, using a 1kHz test tone from a test CD.
My questions are:
1) Is this the correct method for determining the source output voltage, or am I presenting a load with the DMM that is too much for the laptop to handle. I assume the impedance of the DMM would be on the 100 or so kHz level, so I doubt this would cause any problem.
2) I have always heard that source output voltages are typically assumed to be around 1.5V when designing the gain for a generic DIY amplifier. Am I mistaken in this assumption? If that is a typical assumption, then should I be surprised that my CD player was outputting 2.5V? I have no reference as to the recording levels on the CD, so this could be a variable that is not correctly accounted for.
3) Since the gain of the LM3875 should typically be above 10, from what I've read here, which seems to be a result of the required Gain Bandwidth Product from the National Data Sheet, is the method of using a voltage divider ahead of the volume control attenuator a good practice? At this time, I used the RJM schematic for the non-inverting amp, so the actual gain is closer to 32. I could cut this down and reduce the drop across the voltage divider, but I think it would still be required since I am attempting to keep the output of the amp low. Is there a better suggestion for accomplishing this goal?
Thanks for reading through this long post. I thought the chip amp forum would be the most appropriate place to post, even though it is a multi-faceted question regarding source components, speaker measurement and the chip amp as well.
Thanks in advance for any help which you can provide.
Best Regards,
Sandy.