About a year ago I built a pair of loudspeakers based on a Visaton kit - "La Belle". These speakers have a low sensitivity of 81db and are rated 200WRMS, nominally 4Ohm. I didn't use the pre-assembled crossover and instead built my own, using "better quality" parts with the same ratings as in the Visaton crossover.
My power amplifier is a Perrreaux PMF2150B Mosfet amp which has been extensively upgraded, including the 2 big caps which are now 4 smaller ones that double the storage capacity. The amp was tested at 406WRMS @ 8Ohms single channel driven after the upgrade. The pre-upgrade rating at 4Ohms was 400WRMS both channels driven, with enormous headroom. It is claimed to be able to drive difficult loads. SO, I think the amp should be more than capable of driving the speakers to a decent listening level - theoretically about 100db at my listening position at rated power.
Anyway, recently I played an LP pretty loud. According to the dB meter app on my smartphone, average 85db with 95db peaks. That's as loud as I want to go. I just happened to notice that the front panel of the amp seemed very hot after playing the whole LP, so I measured the heat sinks with an IR thermometer, and they were pushing 60C, which is dangerously hot.
I swapped out the Perreaux for an old pair of mono amps and a different pre-amp. These amps got hot too, but they are rated at less than half the power of the Perreaux, and have less than half the heatsink area, so perhaps not a huge surprise. Both power amps are within a degree or so of each other so it doesn't seem like 1 speaker is to blame.
So, that's the background. I hope it's not too long winded and I haven't lost everyone!
I guess my question is what might make the amp overheat so much?
The sound meter app could be understating the SPL, but even so I should be well within the capability of the Perreaux, if not necessarily the Linx monos. The Perreaux should be able to blow up the speakers before the speakers melt down the amp.
I might have wired the crossover wrong, but wouldn't that be audible? To my ears the speakers sound good, and I haven't noticed a change in the sound recently, except maybe that I have had to wind up the volume more recently, but I might be wrong about that.
All of the individual speaker drivers (2 woofers, a mid and a tweeter) seem to be behaving. At the sort of volume I had them playing at, the woofers were moving a lot - not sure how close to their excursion limit, but a lot. The bass is rock solid and no frequency particularly stands out.
I am going to book the amp in to have it checked over, but I have no idea how I would check the speakers. They weigh 40Kg each so pretty hard to move if I wanted someone with more knowledge to do it. If it's my crossovers then it's both, and if it's something like a solder joint, then wouldn't it show up in the sound?
I have tested both speakers individually with a white noise test tone and according to my spectrum analyser app the response is almost dead flat from 50Hz to 15kHz, when it ticks up about 8 dB to 20-ish. In fact, and I find this surprising, it's almost identical to Visaton's published frequency response graph except for that last bit. The published response dips about 3dB at 12kHz, then rises 5db to about 18kHz. I'm under no illusions about the accuracy of my phone's mic, and in any case, there is nothing particularly strange about graphs with real music as opposed to white noise.
So, i hope someone has some insights that they wouldn't mind sharing with me. Maybe this is normal?
My power amplifier is a Perrreaux PMF2150B Mosfet amp which has been extensively upgraded, including the 2 big caps which are now 4 smaller ones that double the storage capacity. The amp was tested at 406WRMS @ 8Ohms single channel driven after the upgrade. The pre-upgrade rating at 4Ohms was 400WRMS both channels driven, with enormous headroom. It is claimed to be able to drive difficult loads. SO, I think the amp should be more than capable of driving the speakers to a decent listening level - theoretically about 100db at my listening position at rated power.
Anyway, recently I played an LP pretty loud. According to the dB meter app on my smartphone, average 85db with 95db peaks. That's as loud as I want to go. I just happened to notice that the front panel of the amp seemed very hot after playing the whole LP, so I measured the heat sinks with an IR thermometer, and they were pushing 60C, which is dangerously hot.
I swapped out the Perreaux for an old pair of mono amps and a different pre-amp. These amps got hot too, but they are rated at less than half the power of the Perreaux, and have less than half the heatsink area, so perhaps not a huge surprise. Both power amps are within a degree or so of each other so it doesn't seem like 1 speaker is to blame.
So, that's the background. I hope it's not too long winded and I haven't lost everyone!
I guess my question is what might make the amp overheat so much?
The sound meter app could be understating the SPL, but even so I should be well within the capability of the Perreaux, if not necessarily the Linx monos. The Perreaux should be able to blow up the speakers before the speakers melt down the amp.
I might have wired the crossover wrong, but wouldn't that be audible? To my ears the speakers sound good, and I haven't noticed a change in the sound recently, except maybe that I have had to wind up the volume more recently, but I might be wrong about that.
All of the individual speaker drivers (2 woofers, a mid and a tweeter) seem to be behaving. At the sort of volume I had them playing at, the woofers were moving a lot - not sure how close to their excursion limit, but a lot. The bass is rock solid and no frequency particularly stands out.
I am going to book the amp in to have it checked over, but I have no idea how I would check the speakers. They weigh 40Kg each so pretty hard to move if I wanted someone with more knowledge to do it. If it's my crossovers then it's both, and if it's something like a solder joint, then wouldn't it show up in the sound?
I have tested both speakers individually with a white noise test tone and according to my spectrum analyser app the response is almost dead flat from 50Hz to 15kHz, when it ticks up about 8 dB to 20-ish. In fact, and I find this surprising, it's almost identical to Visaton's published frequency response graph except for that last bit. The published response dips about 3dB at 12kHz, then rises 5db to about 18kHz. I'm under no illusions about the accuracy of my phone's mic, and in any case, there is nothing particularly strange about graphs with real music as opposed to white noise.
So, i hope someone has some insights that they wouldn't mind sharing with me. Maybe this is normal?
Thanks. That is the circuit diagram that I used. As I wrote, all sounds good, which leads me to believe I implemented it correctly. The white line in my attached image is simply background noise after I turned the music off.
Hello
It's normal for the amp to get hot when working on a 4 Ohm load, especially under heavy use!
After you have to check the good ventilation of the amps! Or make it work on 8 Ohms which requires less the amp.
Or modify the operating point of the amplifier by reducing the bias of the output stages!
You can check the current draw of the speaker by putting a 0.5 ohm resistor in series with it. Be sure the wiring is secure enough to not short your amplifier. For a 200 w 4 ohm speaker the resistor should be rated at least 70 watts. I=V/0.5 You had best measure with an analog meter with 25 or 50 VAC scale, else an oscilloscope. DVM AC scale lie a lot on music, except the $160 RMS model. Use a broad spectrum music source, not a single sine wave tone. More than 7.1 A, you have a speaker problem.
In any case 60°C isn't yet dangerous. 70/75 °C may be risky. The amp isn't a microprocessor. And MOSFET have positive tempco so no risk to thermal runaway.
Thank you for your reply. I'm going to expose my ignorance, but do you mean I should put a 0.5 ohm resistor between the end of the speaker wire and the binding post, then measure the voltage drop across the resistor?
Yes. Between the amplifier output and the speaker hot input. Alligator clip leads with insulated rubber boots may be useful to do this task.