I don't need a higher room.
Yes, one can go well below 300 Hz or so using the conventional gated impulse response measurement techniques, but you need to install the speaker on a large tower in a huge room (way larger what you and I have at home), or you need to go outdoors. There are many descriptions of this out there. A quick Google search brought up this document, which describes the problem (and a possible way out using the beam-forming technique). Take a look at Fig. 9, which shows the first room echo about 5 ms after the anechoic impulse. After gating, the frequency response is good down to about 200-300 Hz, and that's also the frequency resolution (Fig. 10).
It's hard to tell the precise frequency resolution in post 111, but from eyeballing the wiggles in the curve I'd estimate it to 20 Hz. This would correspond to an anechoic impulse-response length of 1/20 Hz = 50 ms, or about 9 m distance of the driver+microphone from the nearest reflective surface. That would imply a room of at least 18 x 18 x 18 m size, with the driver and microphone mounted on 9 m high stands.
@trosse1 Can you share the raw data of your impulse response measurement so we can have a look at this?
Last edited:
You definitely don't know my home 
With 2m microphone height you already get 9-10ms (which is still not enough in my opinion but ok for tweeter measurements). 3m gives 15ms, with 4m you are good to also measure difraction effects of larger speakers. Outdoor totally doable with large stands - but no fun to be honest.
I just got a stable winding speaker stand with 2,4m height - speaker comes on top of that. And I have an unheated hall but with >8m height.
One of the nice things when reflectins are far - they go down in level. At some point you don't have to gate them out. So I put 1m absorption at the reflection point with my stand and end with very little influence from the environment.
Acoustic measurements is not about putting a mic in front of a speaker and do some simulations. Getting GOOD data is the key.
With 2m microphone height you already get 9-10ms (which is still not enough in my opinion but ok for tweeter measurements). 3m gives 15ms, with 4m you are good to also measure difraction effects of larger speakers. Outdoor totally doable with large stands - but no fun to be honest.
I just got a stable winding speaker stand with 2,4m height - speaker comes on top of that. And I have an unheated hall but with >8m height.
One of the nice things when reflectins are far - they go down in level. At some point you don't have to gate them out. So I put 1m absorption at the reflection point with my stand and end with very little influence from the environment.
Acoustic measurements is not about putting a mic in front of a speaker and do some simulations. Getting GOOD data is the key.
Bliesma M74 for €300,-? Don't remember such a price, I thought it was €700,- for 2 from the beginning? But yes, they raised prices and a lot more for the B versions.
These are totally different chassis. The Purify has >10dB less sensitivity and 3-4x the moving mass (the mid version, the x version is even more extreme). 3" voice coil vs 30mm. It's race car vs mercedes luxury car. Both cool but very different.
(car comparisons ... I think that's my first! When you like them I can bring a lot more )
These are totally different chassis. The Purify has >10dB less sensitivity and 3-4x the moving mass (the mid version, the x version is even more extreme). 3" voice coil vs 30mm. It's race car vs mercedes luxury car. Both cool but very different.
(car comparisons
... I think that's my first! When you like them I can bring a lot more )I seem to recall that that the lower cost M74s (A, S and P) were comparably priced to the PTT4.0 around the beginning of 2023. Maybe some of what I noticed was exchange rates at the time.
I think your analogy is apt and actually explains my dilemma to a large degree. I'm not sure my anticipated design needs either a race car or a luxury car or, if it does, which would be the best way to go. And, short of purchasing a pair of each, I seem to have no way of knowing. Then, there's the question of if I should go with M74A, S, or P (I can't afford the M74B) if I went in that direction.
My design will utilize T25Bs, PTT6.5X-04s and CSS SDX12s with a MiniDSP Flex 8 for crossover and DSP (all of that has already been purchased). If I add a midrange between the T25Bs and PTT6.5s, I will do a passive crossover between the T25Bs and the mids. I'm hoping for high transparency, very low distortion and ability to deal with complex and very dynamic (mostly classical) music.
I think your analogy is apt and actually explains my dilemma to a large degree. I'm not sure my anticipated design needs either a race car or a luxury car or, if it does, which would be the best way to go. And, short of purchasing a pair of each, I seem to have no way of knowing. Then, there's the question of if I should go with M74A, S, or P (I can't afford the M74B) if I went in that direction.
My design will utilize T25Bs, PTT6.5X-04s and CSS SDX12s with a MiniDSP Flex 8 for crossover and DSP (all of that has already been purchased). If I add a midrange between the T25Bs and PTT6.5s, I will do a passive crossover between the T25Bs and the mids. I'm hoping for high transparency, very low distortion and ability to deal with complex and very dynamic (mostly classical) music.
I see. Both would fit.
Would the PTT4.0 be loud enough?
What would allow closer acoustic centers?
What membrane material is the PTT6.5?
What amplifiers do you use? Why a passive crossover, you have 8 channels DSP, that fits?
I would say both chassis shuould deliver great transparency and very low distortion. The mid dome is able to bring more dynamic but I don't know from which level on this would be a benefit.
A complete different solution - T34B. Maybe even with a waveguide, you can cross over deep enough for a 6".
I never compared to T25B, I only know T34A which has unreal wide dispersion but is not on the level of the B's in natural representation of highest frequencies.
Would the PTT4.0 be loud enough?
What would allow closer acoustic centers?
What membrane material is the PTT6.5?
What amplifiers do you use? Why a passive crossover, you have 8 channels DSP, that fits?
I would say both chassis shuould deliver great transparency and very low distortion. The mid dome is able to bring more dynamic but I don't know from which level on this would be a benefit.
A complete different solution - T34B. Maybe even with a waveguide, you can cross over deep enough for a 6".
I never compared to T25B, I only know T34A which has unreal wide dispersion but is not on the level of the B's in natural representation of highest frequencies.
I forgot the upper bass drivers. I have 8 GRS 10SW-4s (4 per side) that I'm planning to implement as a slot loaded open baffle. The PTT6.5 (which is a paper composite) will also be open baffle, as would the PTT4.0 if I went that route.
I don't play my music super loud, but the speaker will need to maintain transparency and accuracy across a wide dynamic range. I listen to a lot of Mahler, for example. It might go up to 100 dB, but probably only briefly. Both the T25Bs and any M74 will need to be padded down a good bit to match the PTT6.5 and SDX12s. I suppose that would make them even cleaner. I think the PTT4.0 would be fine. I'll have a good bit of leeway as to where to cross it over to the PTT6.5, so excursion stress can be handled. Still, an advantage for the M74, I guess.
If I used an upper mid, I'd probably cross it over to the T25Bs around 3K - based on distortion and wanting to maximize the beryllium sound characteristic, so acoustic centers should be okay for both. But, yes, something to consider.
The amp is a 6 channel Buckeye Hypex NCore based amp. Hence the passive crossover. I could have purchased an 8 channel, but I thought I'd rather spend that money on a (better) upper mid and go passive there since it would be cheaper.
Yes, I think the T34B would be a better match directly to the PTT6.5, but that decision has been made. I'm very reluctant to use a waveguide with a beryllium driver due to the risk of dome breakage.
If I decided to add the M74, I'd probably go with either the A or the P. Regrettably, I just can't justify spending twice as much on the M74B. The A seems to have better distortion and, possibly, dispersion, but has been described as a bit threadbare harmonically. I'm not sure how noticeable these characteristics would be for me, but that's a tough trade-off. I've wondered if the harmonic weakness of the M74A might be mitigated to some degree by having the T25B handling the higher frequencies.
So you mix open baffle in the midrange with back closed tweeters? Why? How does this sound off axis?
I cross my T25 way lower to avoid to much off axis influences with distant acoustic centers. Doing this critical crossover analog and all the others digital is far from ideal. Selling T25B should be no problem and when you calculate everything you would very likely save money with switching to T34B and save a complete way of your design. No additional crossover will also help the design for sure. Don't save on the wrong end - it will be expensive in the long run! You need the right chassis to make the design work.
M74A shows a lot of deatails in the music but I could not detect any problems with harmonic integration? Where did you read that, in which context? I listened to it active crossed over to T25A or T25B around 2kHz.
A friend uses M74P and is very happy with it. I could imagine it will fit great to the PTT6.5 but also the A version will make no problems.
I only did a listening comparison from M74A and M74S and would say the fabric version does a little "smoothing" of the audio. Great for living room music enjoying, but not for my monitor design. Could not compare the paper version.
I cross my T25 way lower to avoid to much off axis influences with distant acoustic centers. Doing this critical crossover analog and all the others digital is far from ideal. Selling T25B should be no problem and when you calculate everything you would very likely save money with switching to T34B and save a complete way of your design. No additional crossover will also help the design for sure. Don't save on the wrong end - it will be expensive in the long run! You need the right chassis to make the design work.
M74A shows a lot of deatails in the music but I could not detect any problems with harmonic integration? Where did you read that, in which context? I listened to it active crossed over to T25A or T25B around 2kHz.
A friend uses M74P and is very happy with it. I could imagine it will fit great to the PTT6.5 but also the A version will make no problems.
I only did a listening comparison from M74A and M74S and would say the fabric version does a little "smoothing" of the audio. Great for living room music enjoying, but not for my monitor design. Could not compare the paper version.
Most likely "timbre poverty" from the hificompass review, it is balanced with other comments later on
"As with any other speakers with aluminum membranes, there was some timbre poverty, a lack of halftones, and some dryness. Vocals sounded plain and uninvolving, their emotionality was poorly conveyed. Music is not only a set of sounds, but also pauses between them. These pauses or otherwise energy gaps between sounds, were of insufficient depth in very complex music. Sometimes there was a feeling that during pauses sounds do not have time to fade out completely and therefore formed a certain threshold background, that reduced the dynamic contrast of the musical work. There was not enough "blackness". "
I can't do anything at all with such descriptions. Has anyone ever thought about how exactly the midrange drivers have to be integrated so that they all have the same conditions? Were they all midrange drivers of otherwise identical loudspeakers, of course with respective precisely matched crossovers? Did the test listener know which midrange driver he was listening to?
Questions about questions.....
With kind regards
Michael
When I remember right he listens to the speaker wideband? So really only the dome and just put music on.
That's pretty far from any usecase, I cross them at about 2kHz. When listened fullrange you have the resonance and probably some other stuff going on. I would not take that to serious.
I made an prototype just for comparing and listening to these domes. 2 domes side by side and switchable. 2 setups on the DSP to get frequencyresponse matched. A 2nd person switching.
I belive I could hear a difference between A and S, looking forward to my comparisson A and B - not expecting something to hear to be honest when used to 2kHz but let's see.
I appreciate your points, don't shoot the messenger
They are not my opinions just pointing out where Paul most likely got his worry from.But on the flip side hificompass tests a lot of drivers and has quite a few well thought out builds so I don't discount his subjective opinion either, just put it in the pot will all the other available information before making any decisions.
Open baffle bass (SLOB) and midrange with back closed tweeters, yes. That's just the first design I'll investigate. If it's problematic - and it may well be - I'll pivot and go in a different direction. Maybe SLOB with cardioid midrange and closed tweeters (reminiscent of the Gradient Helsinki, though again I'd really prefer to avoid a waveguide with the beryllium tweeter). Maybe I'll end up getting rid of the SLOB between the SDX12s and the PTT6.5. Who knows. All of that is still up in the air. I need to see what the distortion is like on the SDX12s; that is, how high I can cross them over, before I eliminate the GRS based SLOB.
I'll give some thought to selling the T25s and buying the T34Bs. You're right that it might be the cheapest, and best, in the long run.
fluid guessed correctly that my thoughts on the M74A are based on the hificompass review. Of course, I have no way of validating the comment regarding timbre poverty. I can only try to align my expectations to what is written based on my musical priorities based primarily on my preference for classical music.
I don't understand why doing an analog, passive crossover between a mid and the tweeter and digital, active elsewhere would be a problem (assuming they were done well). Could you please elaborate on your thoughts there?
Yeah, that's got to be close to ideal for comparisons. Do you then just return the drivers you don't choose?
It's not a problem - it's a pity.
You can do way more precise and clean active and you run through the DSP and D/A conversion anyways.
Just for fooling around and testing concepts you could stick with the T25B and test all the other stuff first. Or just get a used pair of T34A to stay price sensitive and have the low crossover frequency (which is more important as the area >10kHz).
Not sure If I posted it already here - this is the Monster I built for comparisons. Switching time must be really short to reliably hear differences.
Swapping speaker parts and knowing what you hear ... is fine for the customer but your brain will hear way more as just the drivers. (but I expect significant differences between a total S and B setup - would love to hear that)
Swapping speaker parts and knowing what you hear ... is fine for the customer but your brain will hear way more as just the drivers. (but I expect significant differences between a total S and B setup - would love to hear that)
This is tested in nearfield in an non environment room - there is barely any room influence. Filter is active 4th order.
You can identify the drivers horizontally so for sure it's no perfect test! But the best I could do with short switching times (1-2s which needs the DSP to switch presets).
You can identify the drivers horizontally so for sure it's no perfect test! But the best I could do with short switching times (1-2s which needs the DSP to switch presets).
I believe the driver will sound more characteristic of the way its implemented (ie. LP xover freq) rather than just based on the diaphragm material itself. The motor design also makes a significant difference in distribution of harmonics, affecting the balance of different order.HD modes.
Even VC former material can make a difference, as would the use of a ferrofluid, but most higher end designs wouldn't use that due to its tendency of reducing low level details and macro dynamics. This is probably the least intrusive thing to SQ, but experienced listeners can hear the effects of ferrofluid and the type of VC former used. Aluminum is said to dampen the most and swallow more details being an effective eddy current brake. Its thermally conductive properties does help the driver handle a bit more power.
I've heard some very smooth Alu/Mg and even Ti domes which were easier on the ears than a few silk models.
As a whole, when used in their pistonic range within VC excursion limits, HF drivers shouldn't exhibit too much of their own signature. Once run into breakup mode closer to its limit, that can change alot and will typically start revealing the diaphragm material to trained ears. Silk domes tend to sound mushy and less detailed at higher volume levels.
Metal domes usually play considerably cleaner at higher SPLs. The dome rigidity has alot to do with that and in the case of silk or textile domes, there is always some amount of wide band softer breakup going on which makes things more smoothed over to the ears over longer periods of time at average levels.
I remember when Focal had their TC90K and TC120K, those were some unforgiving tweeters if crossed incorrectly. They actually sounded more fatiguing in some situations than most metal domes, but could be very precise and clean when used correctly. Composite diaphragm materials can be all.over.the place that way. My ears still prefer Alu/Mg and Be over anything else, but there are a handful of Ti domes which break the rule ie. the TW025A28 Audax gold Ti dome.
Even VC former material can make a difference, as would the use of a ferrofluid, but most higher end designs wouldn't use that due to its tendency of reducing low level details and macro dynamics. This is probably the least intrusive thing to SQ, but experienced listeners can hear the effects of ferrofluid and the type of VC former used. Aluminum is said to dampen the most and swallow more details being an effective eddy current brake. Its thermally conductive properties does help the driver handle a bit more power.
I've heard some very smooth Alu/Mg and even Ti domes which were easier on the ears than a few silk models.
As a whole, when used in their pistonic range within VC excursion limits, HF drivers shouldn't exhibit too much of their own signature. Once run into breakup mode closer to its limit, that can change alot and will typically start revealing the diaphragm material to trained ears. Silk domes tend to sound mushy and less detailed at higher volume levels.
Metal domes usually play considerably cleaner at higher SPLs. The dome rigidity has alot to do with that and in the case of silk or textile domes, there is always some amount of wide band softer breakup going on which makes things more smoothed over to the ears over longer periods of time at average levels.
I remember when Focal had their TC90K and TC120K, those were some unforgiving tweeters if crossed incorrectly. They actually sounded more fatiguing in some situations than most metal domes, but could be very precise and clean when used correctly. Composite diaphragm materials can be all.over.the place that way. My ears still prefer Alu/Mg and Be over anything else, but there are a handful of Ti domes which break the rule ie. the TW025A28 Audax gold Ti dome.