Hi !
I have noticed a strange behavior of my hi-fi set-up.
When I listen to the radio through the FM tuner, preamplifier and power amplifiers get warmer than when I listen to CD/SA-CD/DVD-V or LD. In fact, in the latter case, preamp and power amp barely heat up from ambient temperature, whereas in the former, temperature of the cases is noticeably warm, especially just above the regulated power supply of the preamp.
I should add that when I listen to digital discs after the preamp+amps have got warmer after tuner listening, the case temperature of the downstream equipment cool down.
Any idea which could explain such a behavior ?
I have noticed a strange behavior of my hi-fi set-up.
When I listen to the radio through the FM tuner, preamplifier and power amplifiers get warmer than when I listen to CD/SA-CD/DVD-V or LD. In fact, in the latter case, preamp and power amp barely heat up from ambient temperature, whereas in the former, temperature of the cases is noticeably warm, especially just above the regulated power supply of the preamp.
I should add that when I listen to digital discs after the preamp+amps have got warmer after tuner listening, the case temperature of the downstream equipment cool down.
Any idea which could explain such a behavior ?
Last edited:
Amusing. 🙂
Love the humor.
But, unfortunately, it's nor that. For instance, the 2 power amps are meters away from the hi-fi rack.
Love the humor.
But, unfortunately, it's nor that. For instance, the 2 power amps are meters away from the hi-fi rack.
The 2 power amplifiers are right next to the loudspeakers. Rest of the electronics are on a 6-tier rack with glass shelf. From bottom to top : LD-player, DVD player 1, DVD player 2, CD/SA-CD player, preamplifier and, on the top shelf, the FM tuner.
FM is more compressed than CD, so the average power output is higher. This dissipation difference might be noticeable at higher listening levels.
It may even be that some level of unfiltered RF modulation has made it past the preamp’s input filter, but in such event it should have been attenuated again at the power amplifier’s RF input filter.
Are your preamp and power amp DIY, or store bought? Do you have a >100MHz scope on hand?
It may even be that some level of unfiltered RF modulation has made it past the preamp’s input filter, but in such event it should have been attenuated again at the power amplifier’s RF input filter.
Are your preamp and power amp DIY, or store bought? Do you have a >100MHz scope on hand?
No DIY.
The preamplifier is an old Cabasse (French manufacturer) PST16 from 1970. Refurbished. Meets its specs. No input filter. Rise time with 10 kHz square wave is about 630 ns at usual listening level. No trace of instability on resistive load. The power amps are Cabasse Polaris AM1000s. Much more modern, with input RF input filter implemented with pass-through capacitors. No idea of the cut-off frequency.
I have a 150 MHz scope and a 0.5-500 MHz spectrum analyzer at hand, but no FM RF generator and no experience whatsoever with RF or radio.
The preamplifier is an old Cabasse (French manufacturer) PST16 from 1970. Refurbished. Meets its specs. No input filter. Rise time with 10 kHz square wave is about 630 ns at usual listening level. No trace of instability on resistive load. The power amps are Cabasse Polaris AM1000s. Much more modern, with input RF input filter implemented with pass-through capacitors. No idea of the cut-off frequency.
I have a 150 MHz scope and a 0.5-500 MHz spectrum analyzer at hand, but no FM RF generator and no experience whatsoever with RF or radio.
Last edited:
The tuner my have crappy 19khz suppression so there is some power going there. FM is also compressed more making it sound louder.
Yes, I have thought about FM dynamic compression.
As the preamplifier have adjustable sensitivity on all 6 inputs, I have aligned those in order to get roughly the same loudness with FM radio and (good) discs without changing the main volume control. Could it be that simple an explanation ?
As for pilot suppression, i will check level with a selective voltmeter, if possible. But it will be with actual FM radio stations. Not ideal, i guess.
As the preamplifier have adjustable sensitivity on all 6 inputs, I have aligned those in order to get roughly the same loudness with FM radio and (good) discs without changing the main volume control. Could it be that simple an explanation ?
As for pilot suppression, i will check level with a selective voltmeter, if possible. But it will be with actual FM radio stations. Not ideal, i guess.
Did the pre and power amp always get warmer when listening to the tuner as a source, or is this something recent?
If you want, connect the spectrum analyzer to the output of the preamp while the tuner is playing into it. If there are no high-frequency emissions above the audio band, that will at least be crossed off the list.
If you want, connect the spectrum analyzer to the output of the preamp while the tuner is playing into it. If there are no high-frequency emissions above the audio band, that will at least be crossed off the list.
I will try some measurements today.
I have only put 2+2 together recently, but I think it has always being like that (for years). Just that I have since this days not made any relations between case temp and the fact that it occurs with the tuner only.
I have only put 2+2 together recently, but I think it has always being like that (for years). Just that I have since this days not made any relations between case temp and the fact that it occurs with the tuner only.
Hi. Interesting issue with FM tuner you have. Probably fm stereo decoder contains high level of 19 kHz and 38 kHz output , badly filtered. Do you have ac power meter ? You could measure power consumption of amplifier while fm stereo is playing and while watching power consumption, switch to mono mode, i think you would see some decrease in consumption.
Remember in tape cassettes times i've recorded songs from FM diy tuner and output filtering was weak , and you know what - 19kHz pilot tone was easily recorded and caused to disable noise reduction system while playing back these records !
Remember in tape cassettes times i've recorded songs from FM diy tuner and output filtering was weak , and you know what - 19kHz pilot tone was easily recorded and caused to disable noise reduction system while playing back these records !
Actually the 38 kHz carrier is suppressed at the transmitter and restored at the receiver, that might leak in. There used to be an MPX filter at the audio output of elder tuners (in the cassette recorder era). It was in order to prevent beat with the bias oscillator, which resulted chirps in the recording. Apparently it is not needed in more modern tuners. I think the reason is that cassette recorders went out of fashion. But the 38 kHz subcarrier is still there.
The 19 kHz pilot is still there, but with low amplitude. It could be also causing audible beats. But I think the problem here is the high amplitude 38 kHz switching signal that is not properly filtered.
The 19 kHz pilot is still there, but with low amplitude. It could be also causing audible beats. But I think the problem here is the high amplitude 38 kHz switching signal that is not properly filtered.
Thanks for having reminded us with the work of the MPX encoding, lcsaszar.
As I was stuck in garden work today, I have not been able to hook up the spectrum analyzer, but I made some basic measurements with my trusty HP3581A selective voltmeter.
I dug into the service manual of my tuner (Grundig T-9009, almost the same as the T-9000 for all intent and purposes) and read that the 19 kHz and 38 kHz suppression should respectively be <-55 dB and <-60 dB referred to 1 kHz with an RF input FM modulation of 1 mV/75 ohms and 40 kHz deviation.
I also read that the built-in calibration tone of 400 Hz (actually 389 Hz) should be aligned to the same audio output level that the one produced by the above mentioned RF modulation.
Thus, I assume that checking the 19 kHz and 38 kHz suppression referred to the output level of the calibration tone should give a relatively good assessment of the pilot suppression, even if I do note have an FM signal generator to verify the actual performance neither the correctness of the adjustment of the calibration tone.
I have first measured the absolute voltage level of the cal tone from the fixed outputs of the tuner, which are specified to produce 1.7 V RMS +/ 2 dB with the above mentioned 1 kHz FM modulation signal. I have measured exactly 1.74 V RMS.
Then, I have aligned the needle of the selective voltmeter in relative measurement mode with the 0 dB mark on the galvanometer and I have searched for the 19 kHz and 38 kHz tones with real radio stations, having being careful to tune in a station that emit a signal which do not clip the input of the voltmeter - a classical music radio station.
I have found the pilot tone bang on 19 kHz to be -57.5 dB on the left channel and -61.8 dB on the right channel. I have searched for the 38 kHz tone, but the needle barely showed a smidgen of deviation of 2 to 3 dB in the "unspecified" bottom 90-80 dB range of the galvanometer. In any case, the 38 kHz is thus way below -80 dB.
There was obviously some noise or side-bands above 19 kHz and around 38 kHz, but nothing that made the needle deviate frankly above the -80 dB mark. Apart from that noise or sidebands, there was nothing left to see all the way up to the 50 kHz limit of the selective voltmeter.
So far, so good. To my mind, this preliminary check does not show any abnormality and the tuner seems to be in specs.
Perhaps my concern is due to the much reduced crest factor of FM sound as kouiky has pointed out, after all.
As I was stuck in garden work today, I have not been able to hook up the spectrum analyzer, but I made some basic measurements with my trusty HP3581A selective voltmeter.
I dug into the service manual of my tuner (Grundig T-9009, almost the same as the T-9000 for all intent and purposes) and read that the 19 kHz and 38 kHz suppression should respectively be <-55 dB and <-60 dB referred to 1 kHz with an RF input FM modulation of 1 mV/75 ohms and 40 kHz deviation.
I also read that the built-in calibration tone of 400 Hz (actually 389 Hz) should be aligned to the same audio output level that the one produced by the above mentioned RF modulation.
Thus, I assume that checking the 19 kHz and 38 kHz suppression referred to the output level of the calibration tone should give a relatively good assessment of the pilot suppression, even if I do note have an FM signal generator to verify the actual performance neither the correctness of the adjustment of the calibration tone.
I have first measured the absolute voltage level of the cal tone from the fixed outputs of the tuner, which are specified to produce 1.7 V RMS +/ 2 dB with the above mentioned 1 kHz FM modulation signal. I have measured exactly 1.74 V RMS.
Then, I have aligned the needle of the selective voltmeter in relative measurement mode with the 0 dB mark on the galvanometer and I have searched for the 19 kHz and 38 kHz tones with real radio stations, having being careful to tune in a station that emit a signal which do not clip the input of the voltmeter - a classical music radio station.
I have found the pilot tone bang on 19 kHz to be -57.5 dB on the left channel and -61.8 dB on the right channel. I have searched for the 38 kHz tone, but the needle barely showed a smidgen of deviation of 2 to 3 dB in the "unspecified" bottom 90-80 dB range of the galvanometer. In any case, the 38 kHz is thus way below -80 dB.
There was obviously some noise or side-bands above 19 kHz and around 38 kHz, but nothing that made the needle deviate frankly above the -80 dB mark. Apart from that noise or sidebands, there was nothing left to see all the way up to the 50 kHz limit of the selective voltmeter.
So far, so good. To my mind, this preliminary check does not show any abnormality and the tuner seems to be in specs.
Perhaps my concern is due to the much reduced crest factor of FM sound as kouiky has pointed out, after all.
Last edited:
My phone, Galaxy S10, Has a FFT app and this tone can be seen easily. My new FM output circuit suppresses it about 65 dB but it can still be seen as I boost the treble control.
Thanks for that observation.
By the way, I mistakenly reported my measurements in the above post. The -61.8 dB does correspond to the attenuation of the 19 kHz on the left channel with a weaker classical radio station emission. With the same stronger radio station, the attenuation in the right channel is in fact -64.8 dB. I check left channel with other radio stations and found variation of a few db, + or more generally -, of said attenuation compared to the stronger classical radio station.
To be able to listen to good classical radio stations without touching the volume control from having listened to good CD recordings, I have to use the variable outputs of the tuner with maximum attenuation, which per specifications should give about 300 mV RMS more or less with the cal tone. With the input sensitivity adjustment of the preamp, I give that level a further attenuation to obtain 200 mV RMS output, where with the same volume control setting, a 0 dBFS tone recorded on a test CD will give 2 V RMS. That gives a rough idea of the dynamic compression of FM sound, which, incidentally, remains subjectively less dynamic to listen to than good CD recordings at the same volume control setting.
By the way, I mistakenly reported my measurements in the above post. The -61.8 dB does correspond to the attenuation of the 19 kHz on the left channel with a weaker classical radio station emission. With the same stronger radio station, the attenuation in the right channel is in fact -64.8 dB. I check left channel with other radio stations and found variation of a few db, + or more generally -, of said attenuation compared to the stronger classical radio station.
To be able to listen to good classical radio stations without touching the volume control from having listened to good CD recordings, I have to use the variable outputs of the tuner with maximum attenuation, which per specifications should give about 300 mV RMS more or less with the cal tone. With the input sensitivity adjustment of the preamp, I give that level a further attenuation to obtain 200 mV RMS output, where with the same volume control setting, a 0 dBFS tone recorded on a test CD will give 2 V RMS. That gives a rough idea of the dynamic compression of FM sound, which, incidentally, remains subjectively less dynamic to listen to than good CD recordings at the same volume control setting.
What makes me wonder is the preamp getting warmer! To a certain degree I can understand why a power amp may change its power consumption depending on drive waveforms and power delivery to the speakers, I can't imagne how a preamp can do so. Preamps usually do not deliver significant output power (mostly only voltages with very low current) and mostly run in class A. Why should the total power consumption change with different waveforms? The power consumption normally is independent from the waveforms and mostly stays constant.
Maybe preamp is just getting some heat from amplifier ,if they are stacked together? Or preamplifier is oscillating, but only with FM tuner input selected, or tuner signal input ?? Would be very strange. Signal ground loop ?
Also one stupid idea ,if you would have ac power meter and fm modulator with sd card playback function. Compare power consumption in silence , playback paused , with cd input ,but playback paused too. This would eliminate audio compression related increased consumption.