Let's start with a 50W into 8ohms amplifier fed from a +35Vdc supply.
On sustained sinewave the maximum output is 20Vrms and 28.28Vpk.
If you now test it with a low duty cycle signal to minimise the droop of the smoothing capacitance you may get some of that Vsag back as extra output voltage. Let's assume you get an extra 75% of Vsag, i.e. 3V extra when Vsag could have been 4Vsag
That means the maximum output has become 28.28+3 = 31.28Vpk equivalent to 61W, provided the amplifier and PSU can still deliver the higher current of this short term signal. Not 200W !
So let's back track from that. If the amplifier can output a maximum signal of 31.28Vpk when the load is 8r0 and the signal is low duty cycle, then -20dB is 0.61W into 8r0.
Let's think about you listening at 2.5m with a pair of 0.61W signals going to two 87dB/W @ 1m speakers.
you get -2.1dB from the speaker due to 0.61W
Now increase the level because you have two speaker giving +3dB at the listening position.
And now reduce that because you are out at 2.5m instead of 1m giving -8dB.
Summing all that you have 87dB -2.1dB +3dB -8dB = 79.9dB while listening to average levels of 0.61W.
The amps are capable of accepting +20dB if the signal duration is very short, resulting in a peak listening level of 99.9dB
and 99dB if the signal is a sustained bass signal.
On sustained sinewave the maximum output is 20Vrms and 28.28Vpk.
If you now test it with a low duty cycle signal to minimise the droop of the smoothing capacitance you may get some of that Vsag back as extra output voltage. Let's assume you get an extra 75% of Vsag, i.e. 3V extra when Vsag could have been 4Vsag
That means the maximum output has become 28.28+3 = 31.28Vpk equivalent to 61W, provided the amplifier and PSU can still deliver the higher current of this short term signal. Not 200W !
So let's back track from that. If the amplifier can output a maximum signal of 31.28Vpk when the load is 8r0 and the signal is low duty cycle, then -20dB is 0.61W into 8r0.
Let's think about you listening at 2.5m with a pair of 0.61W signals going to two 87dB/W @ 1m speakers.
you get -2.1dB from the speaker due to 0.61W
Now increase the level because you have two speaker giving +3dB at the listening position.
And now reduce that because you are out at 2.5m instead of 1m giving -8dB.
Summing all that you have 87dB -2.1dB +3dB -8dB = 79.9dB while listening to average levels of 0.61W.
The amps are capable of accepting +20dB if the signal duration is very short, resulting in a peak listening level of 99.9dB
and 99dB if the signal is a sustained bass signal.
Last edited:
No, many here would expect peak listening levels around 105dB and a system that starts to get close to realistic levels would achieve 110dB on the short term transients.
that would need something of the order of 500W into your 87dB speakers.
My room filling amplifiers start at 100W and the desk top active monitors use two 50W amplifiers in each speaker.
that would need something of the order of 500W into your 87dB speakers.
My room filling amplifiers start at 100W and the desk top active monitors use two 50W amplifiers in each speaker.
Last edited:
I see Andrew is doing a fine job of selling high-power amplifiers... 
In reality, how much power you need depends on a lot of things. If you change any of these variables slightly, the required output power changes dramatically due to the logarithms and exponentials involved with dB.
As far as I understand human perception, the perceived loudness is correlated with the peak SPL rather than the average.
My listening room (living room) is an L-shape, about 45 m2. I use a pair of LXmini speakers which have a sensitivity around the 85-87 dB (1W, 1m) mark. I have no issues driving those speakers using four channels of Modulus-86 (±28 V supply) to ear-splitting levels. With your MOD286 Mono builds, you should have quite a bit more power available than I do.
My desktop amp is a 5 W Spud powering some Mark Audio Alpair 7P full rangers.
As for the "how much power" question, I suggest taking a look at the latest instalment in my Taming the LM3886 series: https://www.neurochrome.com/taming-the-lm3886-chip-amplifier/output-power/
Tom
In reality, how much power you need depends on a lot of things. If you change any of these variables slightly, the required output power changes dramatically due to the logarithms and exponentials involved with dB. As far as I understand human perception, the perceived loudness is correlated with the peak SPL rather than the average.
My listening room (living room) is an L-shape, about 45 m2. I use a pair of LXmini speakers which have a sensitivity around the 85-87 dB (1W, 1m) mark. I have no issues driving those speakers using four channels of Modulus-86 (±28 V supply) to ear-splitting levels. With your MOD286 Mono builds, you should have quite a bit more power available than I do.
My desktop amp is a 5 W Spud powering some Mark Audio Alpair 7P full rangers.
As for the "how much power" question, I suggest taking a look at the latest instalment in my Taming the LM3886 series: https://www.neurochrome.com/taming-the-lm3886-chip-amplifier/output-power/
Tom
Last edited:
Hi Tom
At first I was a little frustrated today of not having made these calcs earlier. However, I thought let's just finish those two Mod286 first. Then I will see and if nothing helps I have noticed your announcement at the end of the output-power article and I will regard the current built as my practice piece on the way to the ultimate Neurochrome amp ;-)
At first I was a little frustrated today of not having made these calcs earlier. However, I thought let's just finish those two Mod286 first. Then I will see and if nothing helps I have noticed your announcement at the end of the output-power article and I will regard the current built as my practice piece on the way to the ultimate Neurochrome amp ;-)
One point that Tom didn't make is that the LXmini is biamplified -- DSP-based line-level crossover driving one amp per driver with no crossover between speaker terminals and drivers. Generally this means that higher sound levels can be obtained with lesser amplifier power as each amp is only driving part of the total signal. You appear to be driving a more conventional speaker design which makes your situation different than Tom's.
That said, I believe you'll find the Modulus286 entirely satisfactory. The reality is that most people do not listen at concert levels in their home. If you measure the output of your amp when playing music, you'll be surprised how little power you're using most of the time -- and an area where the Mod286 excels due to its low noise and low distortion at all output levels.
That said, I believe you'll find the Modulus286 entirely satisfactory. The reality is that most people do not listen at concert levels in their home. If you measure the output of your amp when playing music, you'll be surprised how little power you're using most of the time -- and an area where the Mod286 excels due to its low noise and low distortion at all output levels.
Hi guys. Thanks for the encouragement. After I got home today I at first got my daily share of soldering done and then started a little experiment. Although not really statically relevant nor very representative. But I hoped it would get me in the ballpark.
I took my smartphone with an SPL meter app on it (yes I can imagine not really accurate, but quickly to set up). After playing some five random music pieces rather on the louder side the app recorded an SPL averaging about 77dB and peaks of about 90dB.
Going back to the SPL online calc. I figured 77dB with my 87dB speakers come out to be 0.5W - who would have believed it. So I guess I might just be fine as 50W (20dB more) gets me 97dB SPL.
One more thought which I got when reading through Brian's comment. The KEFs R700 can be bi-amped, that is the mid and high can be driven separately from the bass (both with a full range signal). Would running a second pair of mono Mods286 with the input in parallel get me an advantage SPL wise?
I took my smartphone with an SPL meter app on it (yes I can imagine not really accurate, but quickly to set up). After playing some five random music pieces rather on the louder side the app recorded an SPL averaging about 77dB and peaks of about 90dB.
Going back to the SPL online calc. I figured 77dB with my 87dB speakers come out to be 0.5W - who would have believed it. So I guess I might just be fine as 50W (20dB more) gets me 97dB SPL.
One more thought which I got when reading through Brian's comment. The KEFs R700 can be bi-amped, that is the mid and high can be driven separately from the bass (both with a full range signal). Would running a second pair of mono Mods286 with the input in parallel get me an advantage SPL wise?
That's certainly true. That said, the tweeter of the LXmini is dialled down nearly 10 dB to match the sensitivity of the woofer. The LXmini is not an efficient speaker.
I agree.
Also note that a 20 dB crest factor is worst case. 14 dB is closer to average. That brings the peak power requirements down from 100x to 25x the average power.
The final consideration is how the amp/speaker combo sounds during a short clip of a tall peak. If you can't hear the clipping, do you need as much headroom? Of course, it's nice to avoid the clipping all together but that does have some consequences for the amp design.
Tom
changing from 87dB/W @ 1m speakers to 90dB/W @ 1m adds 3dB to your SPL levels and is equivalent to doubling your amplifier power.
Adopting 93dB speakers is equivalent to quadrupling your amplifier power.
That way you get to keep your 50W amplifiers and get adequate loudness.
Bi-amplifying your speakers does not get the SPL increase that going active echieves.
Bi-amplifying reduces the loading on the amplifiers and allows the amplifiers to reproduce the signal more cleanly.
The treble in particular can sound much more clean because the long duration bass notes are presented with a very high impedance load and the amp does not deliver high current for these long duration notes.
... got a good deal done today, all resistors, all smd cap, the diodes and one opamp.... other than that, nothing really exiting to report ... boy you've got to give credit to those who do that professionally... see a pic below... not always pretty, but all tested ok so far ...
Therefore let me share my impressions of the Mod286 boards and documentation from a beginners perspective.
Well as you can observe I am doing it. I never thought I would, but the description is so detailed that it is easy to follow. It also gives you some background for example on power supply selection, optional gain settings, stereo or mono operation setup etc. so you understand somehow what it is all good for.
For the actual act of building, the manual gives you practical assembly hints and how to connect power speakers and input connectors. Also there are pictures showing you on how to wind your own inductors for the Thiele network.
For getting the required parts, the BOM can be ordered directly from Mouser. One just follows the provided link and all the necessary parts show up in your shopping cart. Just hit order.
When I ordered, some the parts seemed to be out of stock. Wrote Tom and within a few hours he gave me an alternative part no. Great service.
Therefore let me share my impressions of the Mod286 boards and documentation from a beginners perspective.
Well as you can observe I am doing it. I never thought I would, but the description is so detailed that it is easy to follow. It also gives you some background for example on power supply selection, optional gain settings, stereo or mono operation setup etc. so you understand somehow what it is all good for.
For the actual act of building, the manual gives you practical assembly hints and how to connect power speakers and input connectors. Also there are pictures showing you on how to wind your own inductors for the Thiele network.
For getting the required parts, the BOM can be ordered directly from Mouser. One just follows the provided link and all the necessary parts show up in your shopping cart. Just hit order.
When I ordered, some the parts seemed to be out of stock. Wrote Tom and within a few hours he gave me an alternative part no. Great service.
Attachments
I think you're mixing up bi-amp and bi-wire. For bi-wire one amp drives the speaker through two sets of cables. One cable for the woofer and one cable for the tweeter (or tweeter+mid). The two cables join at the amplifier's output terminals. That changes nothing from an SPL perspective.
Bi-amping is where two amplifiers are used to drive a 2-way speaker. One amp drives the woofer, the other drives the tweeter. Regardless of whether the crossover is active or passive, each amp will serve part of the frequency spectrum, hence share the power. This means you have the possibility of generating higher SPL as Brian pointed to a few posts back.
Yeah... There's a reason this kind of work is done by pick-and-place machines rather than humans.
I suggest mounting all the passives before you start on the opamps. Work from the least ESD sensitive parts (passives) to the most ESD sensitive parts (opamps).
Thank you. I appreciate your input on the documentation. I'm glad it makes the assembly go easier. That's the intent after all...
Tom
Tom thanks for the assembly hints. I hope I didn't accidentally damage the opamp U1.
As for the bi-amping, here is some more info which might help to clarify what I meant. The speakers I have, are passive full range 3-way speakers with built in passive crossovers. At the connection terminals one can turn two knobs to disconnect the HF from the LF section (see picture below). Currently, it is already setup for bi-amping, where one amp feeds the full frequency range to the HF section and another one also the full frequency range to the LF section (Denon AVR amp, where I don't use all the nine channels and therefore two can be configured in the amp setup for bi-amping). Therefore, the amp does not separate the frequencies, it's entirely done by the passive crossovers inside the speaker cabinets.
If one doesn't want to use the bi-amping feature one has to turn the knobs to link HF and LF and only run one cable from the amp to the speaker.
Again, I wouldn't be able to say if two mono Mod286 per channel, one driving the HF and the other the LF part, would provide more SPL / power.
Thanks for giving this some thought. I really appreciate it.
As for the bi-amping, here is some more info which might help to clarify what I meant. The speakers I have, are passive full range 3-way speakers with built in passive crossovers. At the connection terminals one can turn two knobs to disconnect the HF from the LF section (see picture below). Currently, it is already setup for bi-amping, where one amp feeds the full frequency range to the HF section and another one also the full frequency range to the LF section (Denon AVR amp, where I don't use all the nine channels and therefore two can be configured in the amp setup for bi-amping). Therefore, the amp does not separate the frequencies, it's entirely done by the passive crossovers inside the speaker cabinets.
If one doesn't want to use the bi-amping feature one has to turn the knobs to link HF and LF and only run one cable from the amp to the speaker.
Again, I wouldn't be able to say if two mono Mod286 per channel, one driving the HF and the other the LF part, would provide more SPL / power.
Thanks for giving this some thought. I really appreciate it.
Attachments
Last edited:
Tomchr is using the USA definition of bi-amping.
That is really an active speaker.
The passive bi-amping term used by Brian in the last post is a new name invented to try to avoid the ambiguity brought about by the introduction of the USA definition.
Bi-amping when it first came about was using two amplifiers to drive one speaker.
That is what Harmonic describes in post36. This is not an active speaker.
The confusion arrived in post27 when Brian said
The two sides of the Atlantic use the term bi-amping differently.
That is really an active speaker.
The passive bi-amping term used by Brian in the last post is a new name invented to try to avoid the ambiguity brought about by the introduction of the USA definition.
Bi-amping when it first came about was using two amplifiers to drive one speaker.
That is what Harmonic describes in post36. This is not an active speaker.
The confusion arrived in post27 when Brian said
Brian is describing an active speaker setup but calling it bi-amping. There's the problem.
The two sides of the Atlantic use the term bi-amping differently.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.