No - It is as simple that some speaker driver manufacturers have a "listen and peek" test for their expensive drivers where one person listens for the noises, resonances, subbing voice coils etc. that the standard measurements cannot detect
What standard measurement, I do agree you have to listen sometimes for detection, when i did it , a simple sine sweep would let you hear a lot ...
What is better quality, or better manufacturing process ?
Guido Tent's (+Menno + Peter) triode power amp knocked me off my feet at the triode festival a couple of years ago.
And those had $1 each Russian valves, and ordinary toroidals for both power and output transformer use.
(and I hate neither transistor, nor opamp)
No - It is as simple that some speaker driver manufacturers have a "listen and peek" test for their expensive drivers where one person listens for the noises, resonances, subbing voice coils etc. that the standard measurements cannot detect
You must have a rather cramped idea of what a "standard" measurement is. MLS, Klippel, Heyser impedance plots, sweeps, and more will show up resonances and rattles very clearly. Other than Klippel, this is in the hands of any amateur.
No - It is as simple that some speaker driver manufacturers have a "listen and peek" test for their expensive drivers where one person listens for the noises, resonances, subbing voice coils etc. that the standard measurements cannot detect
A rubbing voice coil makes sounds that you can hear but a microphone can not? I suspect there also are driver manus that do no 100% tests at all, only QC samples.
Sy,
Even with the Klippel there are definite ways to misapply the test equipment. I have followed a thread on another site where there is discussion of the limitations of even this test protocol where there are alignment and other problems with certain speaker tests. I wish there was one great set of test protocols that would cover everything but it seems it is going to take multiple tests and types of test for the foreseeable future at least in regards to loudspeaker testing.
Even with the Klippel there are definite ways to misapply the test equipment. I have followed a thread on another site where there is discussion of the limitations of even this test protocol where there are alignment and other problems with certain speaker tests. I wish there was one great set of test protocols that would cover everything but it seems it is going to take multiple tests and types of test for the foreseeable future at least in regards to loudspeaker testing.
What normally happens are that the alignment of the surround and the alignment of the spider and alignment of the voice coil together create a nonlinear pattern and it might only be one frequency that creates a problem when the different forces interact and if measured it may be within some preset limit deviation setting that does not recognize such issues.
I know for sure that when some driver manufacturers finishes a new driver they measure 100 pcs of the first product batch and stores the results.
Then those results are used to calculate the allowed spread in measurements for testing of this specific driver. Thus it must be performed a manual fidelity check to be sure it sounds ok as the testing will not reveal all issues.
John,
analyzing the simplified circuit you posted and letting Q1 and Q2 matched, R2 = R5 = R, gm1 = gm2 = gm, we reached a total gain of [(gm.R) ^ 2] / (1 + gm.R1). Knowing that we can match the currents of Q1 and Q2 slightly modifying the values of R6 or R7 and knowing also that close to the Idss the gm increase is higher, I ask you : the difference between the current of Q1 and Q2 (13mA and 12mA, respectively) causes some significant prejudice regarding of distortion compared to a situation of equal currents?
analyzing the simplified circuit you posted and letting Q1 and Q2 matched, R2 = R5 = R, gm1 = gm2 = gm, we reached a total gain of [(gm.R) ^ 2] / (1 + gm.R1). Knowing that we can match the currents of Q1 and Q2 slightly modifying the values of R6 or R7 and knowing also that close to the Idss the gm increase is higher, I ask you : the difference between the current of Q1 and Q2 (13mA and 12mA, respectively) causes some significant prejudice regarding of distortion compared to a situation of equal currents?
I've mentioned before applying a test signal, for room reverb testing, to cheap speakers - which in fact revealed a resonance problem in the the driver of one channel, this was completely inaudible when playing normal music. Worse, the resonance issue appeared related to the suspension, and slowly lessened to finally disappear over quite a long time period, as the materials fully stabilised.
One wonders how many 'defects' would show up, if one were to luck on precisely the right test scenario to highlight a problem ...
It's OK, I know how people need their comfort zones ... only so much disturbance per day shall be the rule ...
Why not let the OT (Off Topic) issues calm down an let John and the OT (On Topic) get the very first priority. I know some trolls like SY would get a hard time respecting this, but as he have (for the time being) a moderator status he should be the very first to shut up his OT business.
Last edited:
44K plus of on and off topic post and now this .
Please include part 1, much water under the bridge. I'm surprised that anyone would be offended by the observation that both John and Nelson have a posse of followers that treat them as gurus or high priests handing down their knowledge and experience.
I am somtimes reminded of "A Canticle for Leibowitz" when I read these posts repeated often refering to faded old schematics.
A note of irony for Variac, Miller was inspired by his experiences bombing Monte Casino in writing "A Canticle for Leibowitz", I still have my fathers photos of doing the same from a B17.
I just had one of those moments. My TV monitor's processor died sitting there running the volume up then down in an infinite loop. So I put my Roku box into my computer monitor's HDMI input but alas there is no volume control and only a headphone output. I put my headphones onto my cheeks just so they would not be too loud (Sennheiser HD650's) the improvement in sound quality blew me away, I will use them this way from now on.
The devil is in the details.
Excellent observation Scott! I read 'A Canticle...' many years ago, never made this connection. So, even after 44k+ post you come across one now and then that's worthwhile
Jan
- Status
- Not open for further replies.