Nice one @Sentient, watching this with interest.

Forgive my ignorance, but what brand are these?

 

Love the pebble design too.

 

Regards.

Ant.

Nice project ,it might be worthwhile designing as current drive.   https://www.current-drive.info/

On 14/07/2022 at 10:27 AM, BuzzzFuzzz said:

Forgive my ignorance, but what brand are these?

Gday Ant,,, the brand on the speaker plate is "UK Audio Advanced".

    The packing boxes say, made in China.

 

My googling reveals very little about the brand.

 

Just some cheaply made Chinese speakers, that were a good option for this experiment. 😍

Edited July 15, 20223 yr by Sentient

23 hours ago, stereo coffee said:

Nice project ,it might be worthwhile designing as current drive.  

I'll have a look at that Coffee, cheers.

2nd swamp has arrived.

 

And going 4 way active necessitated an amp change.   In my room of unused (hoarded) hifi toys, was an old 5 channel Rotel, that has not been in use for well over 10 years, maybe closer to 15.  I'd forgotten how heavy this beast is !!

 

So we now have....

Yamaha MX-35, 4 channels driving the tweeter and mid's.

Rotel RB-985MkII, 4 channels driving the bass and midbass boxes (1 channel spare).

 

all speaker boxes now converted to direct wiring,,, cheap ass passive components have been bypassed.

 

4 way active.  One electronic crossover per side.

 

the sound is clear, clean, fast, dynamic. 😍

 

 

Edited July 15, 20223 yr by Sentient

30 minutes ago, Sentient said:

all speaker boxes now converted to direct wiring,,, cheap ass passive components have been bypassed.

 

4 way active.  One electronic crossover per side.

 

the sound is clear, clean, fast, dynamic. 😍

 

As is always the case with active spkrs!  

 

3 minutes ago, andyr said:

As is always the case with active spkrs!  

   

Digital xover vs Electronic xover,,, how big is the difference?

 

I dont know.   But this is a simple solution, and the Swamp's seem to do the job quite well.  I'm pleasantly surprised.

 

There are real gains to be had, using electronic crossovers. IMO

On 9/7/2022 at 10:49 AM, Sentient said:

and tweeter looks to kick in at 6k.

That's extremely high

40 minutes ago, Sentient said:

Digital xover vs Electronic xover,,, how big is the difference?

Assuming you use them to make the same filter shapes, zero.

 

The same goes for passive crossovers.... assuming you make the same filter response (as some other sort of crossover) they will (with a few small caveats) all sound the same.

 

The sound is dependant on the filter shapes you make.... no how you make them.

 

That being said.... you can make filter shapes with (some) electronic crossover/EQ devices, that are impossible to impractical to make with a passive crossover.

 

40 minutes ago, Sentient said:

There are real gains to be had, using electronic crossovers. IMO

Yes.

 

... but if you were to take the time to measure and replicate the identical functions of the original passive filters  (using a device which was flexible enough to reproduce the exact filter shape) ..... they would sound identical or very close.    (Yes. Even with the poverty grade crossover components I saw in the photo you posted).

 

If you have a better sound now (and I'm sure you might) then it's due to the better filter shapes, and driver levels, vs the original.

40 minutes ago, davewantsmoore said:

.. but if you were to take the time to measure and replicate the identical functions of the original passive filters  (using a device which was flexible enough to reproduce the exact filter shape) ..... they would sound identical or very close.    (Yes. Even with the poverty grade crossover components I saw in the photo you posted).

 

In my experience, quality components (resistors and capacitors) can make a tangible difference.   no doubt the ones that were in the speakers were poverty grade,, and the filter slopes deployed were radically different.

 

40 minutes ago, davewantsmoore said:

If you have a better sound now (and I'm sure you might) then it's due to the better filter shapes, and driver levels, vs the original.

 

I do have better sound now, on these speakers, definitely.

 

perhaps, but Dave, it's an over simplification,,, and that surprises me, because it's not like you to oversimplify 

 

An active implementation reduces the frequency range (per channel) before the power amplification stage.    

 

A passive implementation filters the signal after it is amplified.

 

And that, is a significant difference in implementation approaches.

 

Doing active, I need more amp channels, but less is asked of each amp.

 

And filtering before the amps, could be argued as, resulting in less signal distortion.   Rather than trying to passively filter a single high powered signal, coming in for multiple drivers (of differing types and needs). 

Edited July 15, 20223 yr by Sentient

3 hours ago, Sentient said:

 Digital xover vs Electronic xover,,, how big is the difference?

 

Perhaps you mean digital active XO vs. analogue active XO?

 

Given that a digital active XO probably requires an A2D conversion before it ... I would say 'folk lore' would suggest that an analogue active XO will at least sound 'cleaner' than a digital XO (given the latter requires a conversion).  But if you're using a natively-digital source ... then this proviso goes away.  And the flexibility, delay & EQ that (generally) come with a digital XO ... deliver significant sonic advantages.

 

2 hours ago, davewantsmoore said:

If you have a better sound now (and I'm sure you might) then it's due to the better filter shapes, and driver levels, vs the original.

 

2 hours ago, Sentient said:

I do have better sound now, on these speakers, definitely.

 

perhaps, but Dave, it's an over simplification,,, and that surprises me, because it's not like you to oversimplify 

 

An active implementation reduces the frequency range (per channel) before the power amplification stage.    

 

A passive implementation filters the signal after it is amplified.

 

And that, is a significant difference in implementation approaches.

 

Doing active, I need more amp channels, but less is asked of each amp.

 

And filtering before the amps, could be argued as, resulting in less signal distortion.   Rather than trying to passively filter a single high powered signal, coming in for multiple drivers (of differing types and needs). 

 

I agree with you, S - IMO, Dave is coming from a simple 'EE' PoV.

 

In addition to all that you listed, I suggest that, as different film capacitors have a different sound - even if they are the same value ... it stands to reason that an active system (as it results in there being no caps or inductors between each amp and the driver attached to it) has the benefit of removing anything that can influence the sound of the amp from the signal path.

 

This is audible as more dynamics ... more resolution.

 

2 hours ago, Sentient said:

An active implementation reduces the frequency range (per channel) before the power amplification stage.    

A passive implementation filters the signal after it is amplified.

This really doesn't affect sound quality .... with some small caveats (that are not usually invoked).

 

2 hours ago, Sentient said:

less is asked of each amp.

This doesn't affect sound quality.... unless the amplifiers (running a full signal range/power) were misbehaving.

 

2 hours ago, Sentient said:

And filtering before the amps, could be argued as, resulting in less signal distortion.

No.... because when you examine the signal, you see that it typically doesn't have less distortion.

 

If it has linear distortion (ie. different frequency response) ... then this is the textbook example of "different filter shape".

 

If it has non linear distortion.... then either the amplifiers or the passive components (super unlikely) were distorting, and they were poorly specced.   ie. the passive speaker was flawed (most aren't)

 

 

To be clear, I'm the biggest proponent of "active speakers", I've ever met ... but there's a whole laundry list of reasons "why active" that many people talk about that are somewhere between "doesn't really happen in practise for a decent passive speaker" .... to "borderline rubbish".

 

The key reasons are so that you can design amplifier gain to match driver voltage and current sensitivity ..... and so that you can filter before the amplifier.....  but the reason to filter before the amplifier is that this gives you the opportunity to create filter shapes that are otherwise impossible (impractical) with filters after the amplifier.

 

I'm the extreme case, filters after the amplifier won't be able to handle the current flow, but this is really only subwoofers, or very very high SPL audio.

39 minutes ago, andyr said:

Dave is coming from a simple 'EE' PoV.

No.

 

39 minutes ago, andyr said:

This is audible as more dynamics ... more resolution.

I challenge you (and anyone) to actually closely calibrate a passive speaker, and the "active version" (like within fractions of a dB at every frequency, for every individual driver) .... and then live switch between them while listening to music.

11 minutes ago, davewantsmoore said:

This doesn't affect sound quality.... unless the amplifiers (running a full signal range/power) were misbehaving.

I don’t think this is a trivial point. I would suggest that in many or even most cases, full signal amps driving average sensitivity speakers, are underpowered and therefore clipping. Taking the load off an amp via active crossovers will almost certainly eliminate this. 

1 minute ago, sir sanders zingmore said:

I don’t think this is a trivial point.

Yes.  It isn't.  Amps sound bad when misbehaving.

 

1 minute ago, sir sanders zingmore said:

I would suggest that in many or even most cases, full signal amps driving average sensitivity speakers, are underpowered and therefore clipping.

I think "most" cases.... it a big stretch....  but it for sure depends on the SPL and the V and I draw of the speaker (including any reactive dissipation).

 

... but it's quite trivial to test for.   We can know it's happening.   Nobody with the ability to test, should ever assume it is happening.... and without observing it, we should probably assume it is not.

 

1 minute ago, sir sanders zingmore said:

Taking the load off an amp via active crossovers will almost certainly eliminate this. 

The point is less that "this will never happen"..... and more that,  this is a simple enough issue to fix in the passive speaker.... and if you just did that... then they would sound the same.

 

There are obviously many corner cases to all of this..... but as an "inherent advantage" of active it's overblown, as you don't need to go active for fix that.

 

The real (super duper dominant) advantage is that you (can) create different filter shapes.... which (can) sound drastically different.

 

 

 

 

2 hours ago, davewantsmoore said:

I challenge you (and anyone) to actually closely calibrate a passive speaker, and the "active version" (like within fractions of a dB at every frequency, for every individual driver) .... and then live switch between them while listening to music.

 

Well, if you can design a 4 way passive crossover network for this speaker system, that closely matches this active implementation,,, I am happy to A/B it blindfolded .   😁

 

I have done this fairly economically, fairly easily, so far.

 

I'm not confident you would achieve results that are indistinguishable, with these speakers, using your own 4 way  passive network design, and using your own 2ch amp costing sub 2k???

15 hours ago, Sentient said:

Well, if you can design a 4 way passive crossover network for this speaker system, that closely matches this active implementation,,, I am happy to A/B it blindfolded .   😁

 

As passive designs typically have less degrees of freedom....... you typically do it the other way around.   Design the best passive speaker you can  (or, use what you have) .... and then design active filters which match the passive networks precisely  (relatively trivial with active filters design).

 

Said another way.... it would likely be impractical or impossible to do what you are suggesting.

 

15 hours ago, Sentient said:

I have done this fairly economically, fairly easily, so far.

👍

 

15 hours ago, Sentient said:

I'm not confident you would achieve results that are indistinguishable, with these speakers, using your own 4 way  passive network design, and using your own 2ch amp costing sub 2k???

 

If we made passive networks.... and then replicated them with active filters.... and used "blameless" amplifiers for everything.

• You would have identical frequency response for every driver
• You would have identical nonlinear distortion for every driver.

The speakers will sound very very much the same.   Identical, save for some small caveats.

 

 

One can say "oh this is just a simple EE perspective".....  and I would muse "what do you actually mean by that?".

 

If it was implying that I'm just reading from a textbook, and have never actually done such (and so that's why I "naively" say what I do) .... then that is incorrect.   I've done it a lot, and mostly from the perspective of "wanting their to be all these magical SQ advantages from active".

 

If it's from the perspective of "speakers don't actually work like the textbooks say they do" .... then this is LOL.

If they sound difference because of differences in nonlinear distortion .... then all bets are off (something went wrong).    Any halfway competent passive speaker has (should have) inaudible nonlinear distortion.    As I said before.... If it doesn't then "going active" fixes the problem with the passive design, but you didn't need to "go active" to fix that - so it doesn't qualify very well as an "inherent advantage" of active. 

 

If they sound different because of differences in linear distortion (different SPL vs frequency) .... then you just didn't correctly adjust one to match the other.    Paradoxically this (as I said before) IS the headline reason why "active sound different" .... because the compared active vs passive will not have the same SPL vs frequency (unless carefully calibrated to).

 

 

 

On 16/07/2022 at 11:23 AM, davewantsmoore said:

........

 

I've quoted you, and snipped everything.

 

You are off on a rant, feeding me nothing but ifs, Dave.

 

You started, by being black and white, that a passive network with the same filter curves, would sound exactly the same.

 

And when called on it, have been back peddling ever since.

 

All I said was, you're over simplifying, and you were.

 

I point out, that multiple drivers, in a speaker are different and have different needs, being one advantage active implementations bring.    And you acknowledge this.

 

All of a sudden, what started as a black and white statement, is now filled with caveats.

 

*shrugs

 

Of course my suggestion is impractical.  I said it, in response, to the farcical suggestion from the outset.

 

Why suggest A/B'ing at all?   That was *your* suggestion.

 

The passive approach is fundamentally different,,, to the active approach.   There are too many variables, to simply put the answer down to crossover alone.  There is no point in suggesting AB comparisons. It's ludicrous and foolish, serving no purpose.

 

All I can say is,,,, in this case, bypassing the passive filters,,, with a 4 way active solution with different filter curves, and amps directly driving the different drivers,,, sounds much better.

 

there is a directness, in the sound, that is quite remarkable.  

 

And it was fairly easy to do, that any humble hifi enthusiast with basic skills could do.  

 

No rubbish. No hocus pocus.  No snake oil. 

Edited July 17, 20223 yr by Sentient

1 hour ago, Sentient said:

I've quoted you, and snipped everything.

 

You are off on a rant, feeding me nothing but ifs, Dave.

 

You started, by being black and white, that a passive network with the same filter curves, would sound exactly the same.

 

And when called on it, have been back peddling ever since.

 

All I said was, you're over simplifying, and you were.

 

I point out, that multiple drivers, in a speaker are different and have different needs, being one advantage active implementations bring.    And you acknowledge this.

 

All of a sudden, what started as a black and white statement, is now filled with caveats.

 

*shrugs

 

Of course my suggestion is impractical.  I said it, in response, to the farcical suggestion from the outset.

 

Why suggest A/B'ing at all?   That was *your* suggestion.

 

The passive approach is fundamentally different,,, to the active approach.   There are too many variables, to simply put the answer down to crossover alone.  There is no point in suggesting AB comparisons. It's ludicrous and foolish, serving no purpose.

 

All I can say is,,,, in this case, bypassing the passive filters,,, with a 4 way active solution with different filter curves, and amps directly driving the different drivers,,, sounds much better.

 

there is a directness, in the sound, that is quite remarkable.  

 

And it was fairly easy to do, that any humble hifi enthusiast with basic skills could do.  

 

No rubbish. No hocus pocus.  No snake oil. 

 

Amen to that, S!  

 

As you said "there is a directness in the sound, that is quite remarkable (compared to the passive setup)".

 

I just cannot understand why Dave has to take issue with that simple statement.  

 

1 hour ago, Sentient said:

You are off on a rant, feeding me nothing but ifs, Dave.

It's a complex topic.... I was only trying to respond to the discussion.

 

There are some theoretical advantages of "active", which when you actually look, are not really there.  They are just not very significant.

 

There are practical advantages of active systems, ability to use the correct gain, and to create arbitrary and precise filter shapes.... which are just enormous.

 

1 hour ago, Sentient said:

You started, by being black and white, that a passive network with the same filter curves, would sound exactly the same.

Correct.

 

1 hour ago, Sentient said:

And when called on it, have been back peddling ever since.

I've done no such thing.  Perhaps you misunderstood what I said, or why.

 

1 hour ago, Sentient said:

All I said was, you're over simplifying, and you were.

No, I don't really think so.....  there are some corner cases, as I said.... but they really end up revolving around "there was something very wrong with the passive system"  (rather, than there being something inherently great about active).

 

1 hour ago, Sentient said:

I point out, that multiple drivers, in a speaker are different and have different needs, being one advantage active implementations bring.    And you acknowledge this.

Only to the degree that they might require different gain / power..... and this allows an active system to use drivers with wildly differing efficiency.   That's something uniquely an active system can do.

 

That's (just one of the reasons) why I said, you would need to "recreate a passive system, using an active system ... and not the other way around (in order to be able to show that they sound identical)"    ['cos natively designed active systems will have filter slopes and gains, which cannot be practical reproduced with a passive networks]

 

(Is that what you were referring to as back peddling?)

 

1 hour ago, Sentient said:

All of a sudden, what started as a black and white statement, is now filled with caveats.

LOL.... perhaps I was getting to far down into the weeds.

 

For clarity:   When the active and passive systems have the same linear and nonlinear distortion ... they'll sound identical.

 

1 hour ago, Sentient said:

Why suggest A/B'ing at all?   That was *your* suggestion.

Why?   (I thought it was obvious)   To demonstrate that what I am saying is correct.

 

1 hour ago, Sentient said:

All I can say is,,,, in this case, bypassing the passive filters,,, with a 4 way active solution with different filter curves, and amps directly driving the different drivers,,, sounds much better.

Awesome.

 

To be clear I was not ever doubting that (or trying to cast doubt on that) ...... I was just talking about why you got the result you did.

 

There are aspects of active speakers that some people would suggest was responsible for the "awesome", which I am trying to tell you are not (responsible).    That's all.

 

Using new driver filters (ie. the linear distortion) is what made the difference..... and when you look at the crossovers in the original speakers (I think we only saw one, right? ....we'll have to imagine what the others were) .... it should be no surprise.

 

I like it.

 

Edited July 17, 20223 yr by davewantsmoore

12 minutes ago, andyr said:

I just cannot understand why Dave has to take issue with that simple statement.  

Where did I do that?.... I find the statement quite believable.

Dave, you're running around in circles.

 

You've just said, that in practise, the filter slopes cant be the same, in a passive network,,, as an active one.   Especially a complex 4 way active system.

 

You just said that.

 

" ['cos natively designed active systems will have filter slopes and gains, which cannot be practical reproduced with a passive networks] "

 

So why suggest that if the filter slopes are the same, they'll sound the same?

 

Why say that, if in practical terms, they cant be reproduced?

 

They are different approaches.  And I am saying, the differences are not *just* in the filter slopes, there are other aspects also in play (and they are not rubbish aspects at all).

Edited July 17, 20223 yr by Sentient

My take:

You can use an active DSP system to do a 2nd order LR XO with no EQ.  You can do the same with a passive XO.  Both speakers will sound very similar.

You can use an active DSP to do an 8th order XO with 10 bands of EQ or an FIR filter set covering the whole range. You can't do this with passive EQ.   The resultant speaker will sound different to the passive XO speaker due to the different filters. 

Does this mean the active speaker is better because it is active or because the XO and filtering are better?  It's a bit chicken-and-egg, isn't it?

 

One question I have relates to the impedance seen by the amp in an active speaker vs a passive.  In the first example could the more complex impedance of the passive XO cause sound differences between the 2?

3 hours ago, RoHo said:

One question I have relates to the impedance seen by the amp in an active speaker vs a passive.  In the first example could the more complex impedance of the passive XO cause sound differences between the 2?

 

In my opinion, definitely.   And it isnt just total impedance load of a passive (3 or 4 way system), and whether an amp can technically drive it.

 

Some very sweet sounding (low powered) amps, can pair easily with virtually any tweeter.    But will struggle to drive a large woofer.   

And some high powered amps will drive large woofers with ease, but dont have the top end sweetness.

 

In my case, because my boxes were physically seperate, i was already running different amps, while in passive mode.

 

So, in this case there are the following aspects in play (on this project):

1. Removing poor quality passive components from the signal path to the drivers.

2. Direct coupling of suitable amps, to the different driver types in the 4 way system.

3. Applying different filter curves and crossover points, using the analogue electronic crossovers.

3b.  Performing the frequency division ahead of the amps, and presenting less information to the amps.

 

 

Now Dave, or anyone else, can argue that 3b isnt relevant.   I feel it is, especially on the low power end of amps (the sweet sounding ones 😍).

 

None of the above, is hot air made up rubbish.  These are all factual differences, that play a role in changing the performance.  Before to After.

 

 

Edited July 18, 20223 yr by Sentient