Are you separating the L and the B in the BL concept correctly?
L is linear usually unless you have a variable wound voice coil.
The B field is the part where you have non linearity. I'm trying to think of how you could back out the B field measurement from the other measurable parameters. It may be possible. But it will need to be a very low voltage levels so it is not exactly useful.
An example. This is not a fully finished design. It is a very long excursion underhung motor. Green is the area that the coil can move within.
The two other pictures are illustrating the B field and the subsequent BL as simulated.
So can you simulate a B field? Yes!
I do not separate B and l.
The issue is that with arguably "same motor", the longer voice coil driver drops Bl faster than the one with shorter voice coil. (So to speak on both +x and -x on average) Once that is the case (I have my reasons and I will ask B&C for more original data), I smell some horsecrap in the mathematical assumptions used to calculate all this in the Klippel method. The first idea is Kms effects leakage to the Bl(x) data.
The issue is that with arguably "same motor", the longer voice coil driver drops Bl faster than the one with shorter voice coil. (So to speak on both +x and -x on average) Once that is the case (I have my reasons and I will ask B&C for more original data), I smell some horsecrap in the mathematical assumptions used to calculate all this in the Klippel method. The first idea is Kms effects leakage to the Bl(x) data.
I think that you have a long way to go before you understand the mechanics of what you are trying to do. And have only clouded concepts of how this actually works.
Hi , are you using multiple segments to be able to optimize change in each part and reduce rendering time & facilitate project carryover or to use different parts (hiperco) for few of the segments.
Yes. It is a very rapid way to make this kind of a motor. When the DXF is transferred to FEMM4.2 it is transferred as a solid piece.
A man is allowed to think that.
My math is rusty for the equations of electromechanical system dynamics, but AI is helping greatly. For the static approach, it is"DOH" trivial.
Though instead of mocking someones dilemma, one could explain how longer coil system has shorter Bl span across X by Klippel data. That would actually be helpful and on topic.
Last edited:
I have requested original B&C Speakers data on 21DS115, but here is the issue depicted:
As can be seen, the 18DS115 with 4mm longer voice coil sees same or steeper fall of the Bl than the 21DS115. Given the "same motor", this does not add up.
It might sound trivial, but "that´s a lost decibel".
Add the ignorance of speaker power curve, that´s another one.
Add less than very optimized port in a bassreflex enclosure, and you can easily see 3dB loss.
These things matter.
As can be seen, the 18DS115 with 4mm longer voice coil sees same or steeper fall of the Bl than the 21DS115. Given the "same motor", this does not add up.
It might sound trivial, but "that´s a lost decibel".
Add the ignorance of speaker power curve, that´s another one.
Add less than very optimized port in a bassreflex enclosure, and you can easily see 3dB loss.
These things matter.
Last edited:
Hi Crash PC - are the following same of the 2 B&C models
1) When you remove the coil and actually measure the Gauss in the magnetic system - one will find one location within the magnetic gap that has the peak Gauss - at that moment if you move the probe even a little bit sometimes a hair , the Gauss will not hold > is this optimum point same for both motor structure , meaning are the motors actually similar.
2) Is there any shorting rings or motor structure linearity treatments that are dissimilar between the drivers.
3) Coil are the coils - absolutely similar - meaning wound with the same wire gauge, insulation and VC height and layers, dia etc.
If all of the above is same - then one of the reason I think maybe that - the suspension system of 18DS115 has a higher restoring force at X +- 3mm, this will be more evident from the physical nature - if you see the movement allowed by a cone 21" surround will be more than a 18inch surround - provided the surround has been formed with the same material - heat, hydraulic pressure and binders. Also most likely a larger spider may have been used in 21 inch - again keeping material , forming pressure heat & binder same for both 18" & 21" spider - the larger spider will allow more travel before exerting - one may put as - max braking.
This can be the likely case - as during spider and cone surround forming mostly the dye (forming mold) is changed - where as hydraulic press pressure - material - heat and binder is kept same and these individual as we say in India - Masala - is not changed mostly in volume/batch production.
These are my thoughts - and this is one of the reason if you crack a diy way to measure these large signal parameters - there will be buyers - when you do put this on group buy, there are may that cannot afford to buy Klippel - and do not have access labs and colleges in their country that have the the money or priority for it either.
Best Regards
Suranjan
1) When you remove the coil and actually measure the Gauss in the magnetic system - one will find one location within the magnetic gap that has the peak Gauss - at that moment if you move the probe even a little bit sometimes a hair , the Gauss will not hold > is this optimum point same for both motor structure , meaning are the motors actually similar.
2) Is there any shorting rings or motor structure linearity treatments that are dissimilar between the drivers.
3) Coil are the coils - absolutely similar - meaning wound with the same wire gauge, insulation and VC height and layers, dia etc.
If all of the above is same - then one of the reason I think maybe that - the suspension system of 18DS115 has a higher restoring force at X +- 3mm, this will be more evident from the physical nature - if you see the movement allowed by a cone 21" surround will be more than a 18inch surround - provided the surround has been formed with the same material - heat, hydraulic pressure and binders. Also most likely a larger spider may have been used in 21 inch - again keeping material , forming pressure heat & binder same for both 18" & 21" spider - the larger spider will allow more travel before exerting - one may put as - max braking.
This can be the likely case - as during spider and cone surround forming mostly the dye (forming mold) is changed - where as hydraulic press pressure - material - heat and binder is kept same and these individual as we say in India - Masala - is not changed mostly in volume/batch production.
These are my thoughts - and this is one of the reason if you crack a diy way to measure these large signal parameters - there will be buyers - when you do put this on group buy, there are may that cannot afford to buy Klippel - and do not have access labs and colleges in their country that have the the money or priority for it either.
Best Regards
Suranjan
Mocking?
I don't see anyone mocking here?
I think the issue is that you seem to have a very strong opinion about certain research.
Which, in itself, is totally fine.
What stands out, though, is that the experience or depth of knowledge behind those opinions sometimes feels limited.
And again, that’s perfectly okay too.
Where it starts to go downhill for me personally is when statements are made as if things must be wrong.
There are so many variables involved in motor design.
Shape, size, spacing, voice coil diameter, and etc etc.
One thing I haven't seen mentioned yet is the stacking density of the windings, which also plays a role.
edit: something @hunter audio just also mentioned right before me 🙂
Restoring force is also a good one as well!
Yet all of this tends to get overly simplified into "same motor and same voice coil length".
Which most certainly doesn't capture the full picture.
Also the above is the core reason - for brain drain, and once a person leaves below 25, they soon forget the part of life they are leaving & adopt to the new - as till 23 / 25 life that was mostly filled with studies and passing exams. It is the later part in life - that one forms a family, own house, holidays and memories and reasons to e rooted to a place.
First I did not mock you. Nor do I wish to. I did give you an example of exactly what you asked for. Your answer was that it's impossible. I will not continue to share and explain the same thing again. Look. Ask questions. Learn from the answers. I gave you graphs that clearly display a forward rearward mismatch in B field strength. It is very common in Loudspeaker motors. Would you like to see a simulation of the Flux field inside the steel and across the voicecoil gap?
Even if everything looks same - a slight change in the composition of top and bottom plate will have effects.
Below is a simulation of a common compression driver motor structure - the field is not symmetrical , to make it symmetrical - lot of corrections are required, and then the goal of 2.2T becomes difficult incorporating the corrections.
K_Audio, below file maybe interesting as a backup process. Crash PC there are some formulas here tt maybe interesting for you.
What? I know it can be wrong on complex questions. I am leading it with just what I need, to write the python code and point me to the approach where acceleration and speed of the cone is used in the calculations. As I said, now it is basically one of the four approaches / datapoints I am following, it does not derermine my total knowledge or way of working.
I have strong SUSPICION. I expressed that the mechanical properties of the driver do not align with the measurement quite a bit, and that it smells to me so much that I decided for attempt to confirm or debunk that with following developments and measurements. Let´s not see what isn´t there. As a non-native speaker, I hardly posses abilities to generate texts to be read in betwen the lines. 🙂
We are talking about Bl here. If restoring force has effect on that, then it perfectly supports my idea of "wrong" in the context.
You are willing to support my claim this way, but not the same claim using my formulation of it.
It is perfectly possible. That´s why we have discussions. I am also perfectly fine with someone handing my own a$$ to me, but it must be on factual terms.
That rather shows misunderstanding of what I´m trying to do.
I am trying to identify, understand and test the discrepancies in the measurement outcomes. The mechanics of some approaches I am taking is straight forward. It is relevance of the gathered data for speaker building that bothers me. I have clouded concept of workings of one of the three data extraction approaches. Your response did not give me what I needed. Not that I demand it, but clubbing me with underlying data with unexplained connection to my suspicion of measurement discrepancy does not solve it. I am sorry for the misunderstanding.
@hunter audio
Thank you for insights.
I did not have that possibility before, but due to the other needs, I am probably borrowing a gauss meter, and cutting one healthy speaker for various resons to measure all proportions and dimensions needed in order to verify my findings, or make more of these based on checking with these physical driver parameters.
Though I will not have this opportunity for these I picked for comparsion - 18DS115 and 21DS115. Cannot afford to recone em due to other needs, hobbies, endeavors etc.
Probably not. I know why you ask... Though that would more affect general Bl level proportionally, falloff not so much. Somehwat, yes, sure.
Yes, I indirectly pointed to that possibility. But we are looking at the Bl curve here. So If we are justifying this issue by suspension, then I already expressed that I am suspecting suspension (Kms(x)) effects leaking into Bl measurements. And to verify or discard my suspicion, I am exploring three different measurement methods of the same thing. I don´t understand why someone (not you) would be bitter about it...
That is not to be cheap either. Depends on how you look at it. If we ignore sound interface and needed amplifier, then it is $400 scope, $150 DMM, $1200 Keyence sensor solution, $300 shipping, taxes tariffs, $500 Program + markup + safety for RMAs. $2600 certainly is not for everyone. If more info is spilled about the Dayton´s approach (how do they get excursion data), then the rumored higher powered version is already a winner.
So I guess it is just me playing in my sandpit. But the data might be valueble.
My concern and preliminary conclusion though is that this underlying data has low usability in real world of subjective feeling about the speaker design or final THD caused by the operation of the speaker and enclosure together. Meaning, flawed enclosure design can ruin even the best speakers. More straight forward markers are THD, SPL efficacy, longetivity of the hardware.
Yes I am not sure if these drivers are as same as needed to be. Just spotted a discrepancy while some generic claims ARE that the motor is basically the same, so I am trying to dig at it. Longer voice coil on the "same motor" having more Bl falloff than shorter coil does not sit with me right, and I would be happy to get any answer on that. Wrong measurement data might not be true, but suspecting it sounds like first healthy shot at diagnostics of the issue.