And it seems that just this morning we did not agree.
I am not the cat's meow when it comes to motion control. But it concerns me that I have to explain to "seasoned" engineers what gain and phase margin are...why a motor magfield force profile is non linear, how the phase margin is force dependent, why insertion of a second order filter in the control stream lowers phase margin and stability.. When they cannot even understand the fundamentals behind the explanation, there is a problem!!
The motion control vendors are dumbing down their tuning teachings to the point that the endpoint product is just garbage. And people wonder why I do not like flying...
Even here, the discussion of amp and speaker impedance...really? What happened to sqr(-1) and calculus???
The previous was not a rant....not even close.
Jn
Well, a lot of undergrads just learn to cram and take tests to pass. I can totally understand why most new grads with a BS wouldn't know a lot about control systems. There is not a lot of opportunity to go in depth. I would expect most candidates with relevant work experience to know something, though.
In audio, it's a different story as there are a lot of self-taught people that never learned the foundations.
The candidates I have interviewed, I could not ask relevant questions because I was not there to embarrass. It was painfully obvious.
One of the interns, I told upper management (in two divisions) they need to hire him next spring when he graduates. As DPH mentioned...trainable, hungry, and motivated. When I point out where he lacks in understanding, and give him ppt presentations I've created for my work here, he sucks it up, learned it, and calls his professors to see if it is part of the curricula, and if not, why?
Jn
One of the interns, I told upper management (in two divisions) they need to hire him next spring when he graduates. As DPH mentioned...trainable, hungry, and motivated. When I point out where he lacks in understanding, and give him ppt presentations I've created for my work here, he sucks it up, learned it, and calls his professors to see if it is part of the curricula, and if not, why?
Jn
A coaster has been proven by accredited wine judge.
Dan.
Oh ok, sorry about getting that wrong. Can you give more info please, PM if you don't want to discuss it here. I have a passive coaster version, I gave my mother one and she guards it with her life. Nowadays I can stick some putty on the bottle or jug and some on the glass or cup and presto different and preferable beverage be it wine or soy milk or juice or water or whatever. Electric coffee machine, food, water, lighting, video, audio, it's a whole lotta interesting fun !.
Thanks Esa.
I have no problem with that. I know what your preference is and I have a different one.
You believe that current driving is the solution, I do not.
But what you have done I believe is helpful in pointing out an alternative, not current driving, but more like constant current with a voltage source.
This has the added benefit of being able to use any source Z, locks in the crossover and the bass alignment. Did you know it locks in the crossover, that was a benefit we discovered and now you can use a current source too.
Now Esa, when we do this parallel, we hear a similar improvement to the one that you hear with current drive. In fact, take an 8 Ohm speaker and add 18 Ohm series and a room full of guys heard the improvement (this demonstration was conducted by Menno Vanderveen at ETF17). But a similar improvement is heard when using a parallel 8 Ohm in parallel. Many have heard this, indeed speaker manufacturers are doing this right now inside their speakers and not telling the end user, hush-hush.
Ask Richard Marsh here on this thread, he has heard it, many have! Richard has even linked it to back-EMF. And I believe he is right.
What is clear is that parallel resistance works better in flattening the current phase angle:
Here is series 18 Ohm, Red is before, Blue is 18 Ohm series:
Here is parallel 8 Ohm added:
The same improvement is heard when equalising the current of the amplifier (passive caps, inductors etc), where the EQ is on the amplifier side and not on the driver side. This is not about the terminal impedance of the driver, this is about making the current phase angle of the [single] driver go as close to zero and the amplifier produces the same current at all frequencies.
This is what we are exploring, note the we.
-----
Re others here: I shall not wake up every morning and find an avalanche... I get the picture of what is going on. Also, there are some of you that I will just not reply to anymore. Unless it is respectful.
You keep saying that I have not posted a measurement. Well, I did.
You ignored it. I then posted it again, and deja vu, you just ignored the measurement again. I am not inclined to do it for the third time.
Oh, the two above? That is a real driver been measured, what did you think it was?
-----
I wonder how many of these ghosts hiding behind pseudonyms have anything like the real world experience that I have.
There is life beyond text books!
Last edited:
Are you thinking musical instruments as well as audio production and reproduction? Thinking offhand, I don't think Sony ever dipped into musical instruments (even counting turntables, Sony wasn't known for TTs anything like the SL-1200 which of course started a new trend in the musical use and abuse of TTs).
I'm very often impressed with many Yamaha musical instruments, as are some famous people. The Internets says they made very few (a single digit quantity!) of this thing:
("Hmm, a string quartet" I thought on first hearing, but then I thought those string players are awfully tight, and I figured out it was some kind of keyboard and "this ain't your father's Mellotron")
YouTube
(This band was part of my introduction to classical music)
YouTube
There's a Yamaha model strat-clone electric guitar that apparently came with a cheap amp much like the Squier ("by Fender") guitar-amp combo, floating around the local makerspace, left by a previous member who moved. It was missing strings AND (I discovered once I strung it up) an output jack. Once I got it going I set the string height and intonation, and was pleasantly surprised at how well it played and sounded.
Then I got this model acoustic for under $100 at a thrift store. It sounded good in the store, but once I got it home and tuned up I was really impressed, and sure I'd got one of the more expensive models, but no, judicious googling found 2010 new street prices of $249 and $199 (but still really glad I got it), and I can only wonder if they discontinued it because it cut into the sales of more expensive models. The price and laminated top both say it should NOT sound this good. I agree with this review and it's not the only one that reads like this.
Amazon.com: Yamaha F345 Sycamore Top Acoustic Guitar Tobacco Brown Sunburst: Musical Instruments
Back to audio, I recall the Yamaha Sound Reinforcement Handbook has been a standard text for selecting and using live sound equipment.
I recall from over 20 years ago, Behringer was good at making products that looked remarkably similar to competitors' products. I'm not saying anything that's not in Wikipedia.Behringer... um.... let me think of something good to say about them.... they were probably solely responsible for keeping the hot melt glue people in business.
Last edited:
Ever since Google released TensorFlow, any damn fool can learn ML from rapping Youtube AI stars.
YouTube
That made my day.
I once asked an EE friend, a mains AC generator, is it a voltage source or current source? He said current source. I paused for effect and asked him, "are you sure?"
I definitely will not give his name because he is on diyaudio.com and he works for Transfield, now Broadspectrum, here in Oz. This is his job that I was asking him about. But he is pretty sharp and a good-natured guy, but well over six foot, so don't cross him, OK?
I met a lot of people who were academically very bright. These are the folks that can retain information easily and do well in exams, but have no real love of the field. When I was in industry a few years back, we hired in a few grads every year. Because we were a big company, we got to pick the best (same with other big MNC’s as well). At least half left the company to pursue fields completely outside their university training. One guy started off doing modeling and did a great job but resigned after a year and went into house building.
A friend’s son has a PhD in Physics (Quantum holography). He’s retrained as a programmer (Python) and now works for one of the investment banks in London.
You have shown this measurement before in your #27689. You have received replies by Pavel #27691 and me #27690. My remark was that I don't know of any amp that distorts less at lower load impedance, Pavel's that this parallel resistor does not change the current flowing through the voice coil. These answers should be sufficient for you to understand that you are barking up the wrong tree.
Indra replied that there is an SET amplifier he knows that distorts less at 4 than at 8 ohms. That was a good find, but not typical.
You, on the other hand, did not respond. At all. You just reposted the same measurements. Now. This gets annoying. The only relevant measurement would be: distortion profile of the acoustic output before and after.
The measurement you now reposted only proves that there apparently also is life before text books.
Last edited:
Now that is a very blanket statement....
Maybe the big company was a lousy employer?
Maybe half of the guys were smart enough to discover more interesting jobs, being able to look beyond what they "learned"?
This thread might benefit from some fresh input.
Sometimes it smells like an old people's nursing home
Last edited:
It is one of several 'planned obsolescence' features built in to most life forms.
The physicist has the advantage of being quite familiar with complex numbers (more so than a typical EE) and also not being fazed by maths, hence he can learn it quite quickly. I don't recall loop stability being part of my physics course, but we did a lot on complex analysis so we knew all about poles and zeros and analytic functions and branch cuts.jneutron said:
More recently (although still some years ago) I did an EE MSc. I was surprised to find that quite a lot of overseas students had never seen Maxwell's equations.
Nowadays you may find EE lecturers who struggle with basic electronics. They may be experts in ergonomics or computer vision or scalable networks but clueless on BJT biasing.RJM1 said:
Irrelevant. The fact that one person had a bright idea which turned out to be right does not mean that everyone who thinks he has a bright idea will turn out to be right. Most are wrong. We tell Joe that his 'ideas' have no merit because they have no merit and appear to be based on inadequate understanding. We would like him to gain understanding but thus far he has proved resistant.john curl said:
'Resistant to understanding'. I like that.
What’s All This Pease Prototype Stuff, Anyhow? | Electronic Design
Not going to comment on if this was 'good old days' or not.
What’s All This Pease Prototype Stuff, Anyhow? | Electronic Design
Not going to comment on if this was 'good old days' or not.
Let's see if I understand the last few months if the offering here that is usually made to the porcelain throne. When you apply a voltage to a load that does not follow Ohm's law the current does not follow the voltage. There is some resistance in series with all practical amplifier-loudspeaker hook-ups. Some of this is internal to all audio amplifiers and other from cabling and even some to a passive crossover network if used. An over simplification is that force on the loudspeaker's cone is completely proportional to the current through the voice coil. When using voltage drive on the non-ohmic load the voltage drop across the ohmic losses results in increased voltage distortion.
When using current drive the ohmic loses do not change the current in the non-ohmic load. In a perfect transducer the cone displacement must increase as the signal frequency decreases to maintain the same energy output. Loudspeakers have what is called a mass breakpoint where with increasing frequency the motor can no longer drive the cone assembly completely. Loudspeakers as commonly used are designed to have either flat on axis response or flat power response when voltage driven.
When using current drive the ohmic loses do not change the current in the non-ohmic load. In a perfect transducer the cone displacement must increase as the signal frequency decreases to maintain the same energy output. Loudspeakers have what is called a mass breakpoint where with increasing frequency the motor can no longer drive the cone assembly completely. Loudspeakers as commonly used are designed to have either flat on axis response or flat power response when voltage driven.
My friend learned it quickly because I am a great teacher!!!!
Seriously though, he is one of the most brilliant guys I know. I do the EE and mech and motion dynamics, he does the analytical math and coding. His first pass code generated accuracy of motion 3 orders of magnitude beyond current state of the art.
As I recall, I only learned and worked Maxwell in an elective course, so most of my peers were not exposed. And the ten or so in the course, only one got it. They hated my guts.. I mean, who could blame them...back then I was young and stupid, I'm older now..
Jn
Being as I have no EE training whatsoever I represent the uneducated masses that just happen to enjoy tinkering with their systems and in my case rely mainly on subjective perception @ lp.....that being said I do believe most issues with designing passive xo’s by ear is the ever changing phase response with each change.
Sounds to me like Joes design basically self levels the phase which would in turn allow crossover changes without having to worry so much about polar response, imaging, etc.?
If it really works that way it would be on my short list.
If I’m confused.......please disregard!
Sounds to me like Joes design basically self levels the phase which would in turn allow crossover changes without having to worry so much about polar response, imaging, etc.?
If it really works that way it would be on my short list.
If I’m confused.......please disregard!
Virtually nothing conforms perfectly! But the obvious one is a diode, more importantly there are resistors that are made with specific characteristics to correct for issues such as the drift in semiconductors. The most common version is the incandescent light bulb often used this way in oscillators to keep the level stable. But I am sure you have run into thermistors that the resistance varies with temperature, available in both positive and negative coefficients. Using them they self heat and change value.
Then when we get into impedance there are fun things like inductors who values vary with current.
They made "advances" in capacitors, too. Those little MLCC capacitors that look like about a 1mm cube and rated at 10uF 10V, well, they're 10uF at 0 volts but are 5uF at 10V (there's a graph of this right there in the datasheet, but who reads that). I'm sure there's some highly degreed EE using those for timing or filtering that's wondering what the heck is going on in their circuit.
I predict more
- Status
- Not open for further replies.