Any form of time alignment is valid only on one exact point in a 3D space with real drivers mounted vertically on a flat baffle, you can choose which. Not always for the tweeter axis as some may be better few degrees off-axis. And obviously once you have drivers with vertical distance between them, the distance from the array means your relative distance from each driver changes and defeats any attempt at time alignment.
Concave line array and coaxial driver seem to be two possibilities to combat this problem (apart from 'fullrange' drivers), if in fact it is a problem. In my experience this is mostly true for midfield distances, in larger spaces the environment has a significant influence on time domain performance. In the nearfield (assuming a small array) there are very few room effects and you mostly are listening to a point source.
Anyway your question was about the F5 and an active 3-way. I built my own 3-ways with Volt B2549, and Satori 5" woofer and the Satori Neo tweeter. We specifically chose a fully passive setup with the top portion of the baffle tilted back (the low crossover is so low that time alignment is not a problem, but this M/T combo does not behave properly on a flat baffle anyway). System nominal impedance is 6 ohm with a low of 4 ohm, sensitivity is 88dB/w, and a two-pair F5 Turbo drives it beautifully and it goes significantly louder than my threshold of hearing will allow. It also has in-room response down to 18Hz and the large driver handles bass like a beast. The F5 is really a great match.
The output devices in the F5 do not sound nice when used at low levels of bias. If you are using low-power F5s, keep the supply voltage as low as possible, so you can run the bias currents high. A baseline is 1.1-1.3A for a single pair or at least 700-800mA/pr for 2 pair output - less than that the sound tends to get muddy. I use 3pr outputs very often and the amps like to run around 2-2.1A bias for a great sound. Similarly, I like to use about 10VA/w of power output, so a 600VA transformer would be typically used for 30W/ch amp.
Basically don't assume you can use same transformer if you need to run lower and higher power amps off the same power supply, or you should plan to use cap multipliers to drop the voltage to the amp for the top/mid. And another note on the active crossover. If you don't follow the exact discipline that Patrick's pdf showed you will never be able to arrive at the correct crossover topology. The only bonus is that you don't have to deal with driver impedance in active setups, but that doesn't mean you can plonk in any slope and frequency you like.
While doing this exercise we inferred that sometimes, a passive setup can outperform an active setup if it is properly done, the only caveat being the unavoidable drop in sensitivity. Good driver pairings have symbiotic rolloff and impedance characteristics, and very good drivers tend to have excellent correlation between data and retail product. This makes them simple to work with. Don't forget that though you will gain back the insertion loss from the woofer inductor, you still have to plan for 2.5-4dB of baffle step compensation. You might know this already, but just putting it out there.
Concave line array and coaxial driver seem to be two possibilities to combat this problem (apart from 'fullrange' drivers), if in fact it is a problem. In my experience this is mostly true for midfield distances, in larger spaces the environment has a significant influence on time domain performance. In the nearfield (assuming a small array) there are very few room effects and you mostly are listening to a point source.
Anyway your question was about the F5 and an active 3-way. I built my own 3-ways with Volt B2549, and Satori 5" woofer and the Satori Neo tweeter. We specifically chose a fully passive setup with the top portion of the baffle tilted back (the low crossover is so low that time alignment is not a problem, but this M/T combo does not behave properly on a flat baffle anyway). System nominal impedance is 6 ohm with a low of 4 ohm, sensitivity is 88dB/w, and a two-pair F5 Turbo drives it beautifully and it goes significantly louder than my threshold of hearing will allow. It also has in-room response down to 18Hz and the large driver handles bass like a beast. The F5 is really a great match.
The output devices in the F5 do not sound nice when used at low levels of bias. If you are using low-power F5s, keep the supply voltage as low as possible, so you can run the bias currents high. A baseline is 1.1-1.3A for a single pair or at least 700-800mA/pr for 2 pair output - less than that the sound tends to get muddy. I use 3pr outputs very often and the amps like to run around 2-2.1A bias for a great sound. Similarly, I like to use about 10VA/w of power output, so a 600VA transformer would be typically used for 30W/ch amp.
Basically don't assume you can use same transformer if you need to run lower and higher power amps off the same power supply, or you should plan to use cap multipliers to drop the voltage to the amp for the top/mid. And another note on the active crossover. If you don't follow the exact discipline that Patrick's pdf showed you will never be able to arrive at the correct crossover topology. The only bonus is that you don't have to deal with driver impedance in active setups, but that doesn't mean you can plonk in any slope and frequency you like.
While doing this exercise we inferred that sometimes, a passive setup can outperform an active setup if it is properly done, the only caveat being the unavoidable drop in sensitivity. Good driver pairings have symbiotic rolloff and impedance characteristics, and very good drivers tend to have excellent correlation between data and retail product. This makes them simple to work with. Don't forget that though you will gain back the insertion loss from the woofer inductor, you still have to plan for 2.5-4dB of baffle step compensation. You might know this already, but just putting it out there.
Hi AntonTR, I have 6 channels of Aleph J driving 2 Woofer + horn MTM speakers. See pics. Sounds great, but too much heat from this 600W electronic heater, so I use Icepower in warmer seasons. You may want to consider it. Also, I think using Class D just for woofer would not be a bad idea at all.
http://www.diyaudio.com/forums/pass-labs/241729-aleph-illustrated-build-guide-174.html#post4641691
http://www.diyaudio.com/forums/pass-labs/241729-aleph-illustrated-build-guide-174.html#post4641691
Hello, plasnu.
I will check your link a little bit later, thanks.
I understand about class D and also consodered class AB for woofers.
But accordong to the box calculation, my preselected woofer in sealed box will reach its Xmax at 21W average power govig 103dB as I remember.
Seems, it does not require class D here at all.
Or I have to replace woofer...?
Room is 28 sq. meters - if it is really not enough?
BTW, I thought about Aleph J but it is single ended, while my mid and twitter are 4 ohm. I need to estimate if there is enough capacity.
Anton
I will check your link a little bit later, thanks.
I understand about class D and also consodered class AB for woofers.
But accordong to the box calculation, my preselected woofer in sealed box will reach its Xmax at 21W average power govig 103dB as I remember.
Seems, it does not require class D here at all.
Or I have to replace woofer...?
Room is 28 sq. meters - if it is really not enough?
BTW, I thought about Aleph J but it is single ended, while my mid and twitter are 4 ohm. I need to estimate if there is enough capacity.
Anton
Last edited:
Hi Anton, What I wanted to say is not to forget having 6 channel Class A amps may dissipate crazy amount of heat even when no sound is coming from your speakers.
BTW, my Class-D is Icepower 50, rated 50W.
I used BASTA! software for preliminary power calculation.
If start from woofer with its Xmax at 21W and 103 dB this means 2.3A of class A into 8 ohm coil.
Just fine for F5 with it's 2.6A push pull stage.
Assuming baffle step, to reach same 103dB the mid driver will require 16W only.
So, it will need 2 A in class A into its 4 ohm coil.
F5 is also OK here, right?
Seems, twitter will require 8-10 W only.
If start from woofer with its Xmax at 21W and 103 dB this means 2.3A of class A into 8 ohm coil.
Just fine for F5 with it's 2.6A push pull stage.
Assuming baffle step, to reach same 103dB the mid driver will require 16W only.
So, it will need 2 A in class A into its 4 ohm coil.
F5 is also OK here, right?
Seems, twitter will require 8-10 W only.
If you want to have single ended amp for mid or high, you may want to check MOFO by MR. Low dissipation, high damping factor, PSU is laptop PSU!
Oh, I was not recommending particular class D, but Icepower is not a bad choice if you can get it cheaply on Ebay. I paid @$60 par stereo board, and it includes PSU.
I think regular Aleph J dissipate a bit more than regular F5, but not much difference, right?
AntonTR, I'm not trying to confuse you, sorry, but dissipation was what I had not seriously considered before building this 6x AJ, so I just wanted to tell you that.
While I'm 120% happy with the sound of this amp, but due to heat in my room, I'm planning to make another 6 channel class-A amp, it probably will be choke loaded single ended like Mofo or L'amp. 1/2 dissipation of AJ.
While I'm 120% happy with the sound of this amp, but due to heat in my room, I'm planning to make another 6 channel class-A amp, it probably will be choke loaded single ended like Mofo or L'amp. 1/2 dissipation of AJ.
It is none of my business what you, AntonTR, are going to build, but making 3 stereo classA amps for 3 way speakers is a total waste. ClassA is great for tweeters and mids, but for woofers its a waste, any class AB or class D amp is just fine. I would even use passive between tweeter and mid and use only one classA amp. But that's just me, I am lazy by nature. Well, not exactly, I have about dozen amplifiers to swap and compare the sound. What I do is active crossover at ~150Hz, then have woofers powered by the same classAB amp, and then keep swapping only stereo classA amps for mids and tweeters. That's enough excitement for me. Cheers!
(I have not read all thread)
(I have not read all thread)
Thank you for you opinion, Mr. Adason!
Seems, this approach is like a golden balance between efforts and result.
Some people have already offered this arrangement.
I think it worth to consider.
So, your crossover has a two way filters sections:
20-150 Hz for woofer and 150Hz high pass for mid and twitter, right? Then you use passives between mid and high driver as usual, correct?
What is the filters order?
Seems, this approach is like a golden balance between efforts and result.
Some people have already offered this arrangement.
I think it worth to consider.
So, your crossover has a two way filters sections:
20-150 Hz for woofer and 150Hz high pass for mid and twitter, right? Then you use passives between mid and high driver as usual, correct?
What is the filters order?
Last edited:
Sissy amp for Sissy speakers
I love Frenchy speakers
they're mad as Hatters
have a pair of Cabasse 8'' FRs from Dinghy , have a pair of 2-way small Epsilons
yeah , they're mad as Hatters , proudly
Class-D just for woofers seems to be a good idea, and seems to be logically correct, but I "personally" prefer that all the drivers are driven by the same amps. When different amps are used at the same time, I feel like comparing amps, and I can't stop ignoring the drawbacks of each amps. In my ears, amps are fighting. I know I should appreciate the strength of each amp, but I'm not that smart enough. Funny.
Anyway, the whole point of active crossovers is to find work for all the
amplifiers we build.
Amen.
My problem is that i run our of the rooms.
Well, there are other benefits, like if your woofers are lower efficiency, you do not need to kill power to midrange and tweeter with resistors, you just adjust level.
Or if your woofer amp is clipping, it does not polute mid and tweeter with upper harmonics of distortion.
Or you want to save on expensive duelund caps or foil inductors.
Or you built too many amps.
Or if your woofer amp is clipping, it does not polute mid and tweeter with upper harmonics of distortion.
Or you want to save on expensive duelund caps or foil inductors.
Or you built too many amps.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.