|22nd July 2008, 10:35 PM||#31|
Join Date: Feb 2001
Location: Columbia, SC
And what did you 'learn' from the simulation? Nothing. The RIAA curve is easily found on the web, so the shape of it shouldn't be a surprise. Did you learn that Nelson's RIAA circuit was wrong? No. Did you learn that the circuit works as designed? Well, yes, but anyone could have told you that ahead of time. Or you should have realized it yourself. Me, I'd rather have spent that time building a circuit I could listen to rather than wasting it on simulations that accomplish nothing.
Hint: If you've got a proven circuit's schematic in hand--and I think Nelson's stuff qualifies--you can probably build it and expect it to function properly.
You'd have to have a pretty screwy RIAA EQ not to come up with a curve that will 'totally look the same' as an RIAA curve. The devil, as always, is in the details. The curve is basically a 6dB/oct low pass with a flattened spot in the middle. But getting it right...that's another matter, entirely.
I don't have an Agilent either, but it doesn't stop me from measuring RIAA accuracy. I don't like RIAA networks as test bench stuff because then you're faced with the errors in the bench network as well as the errors in the phono stage.
True story: Carver put out a preamp a while back. The preamp had a phono stage (optional) but also had a passive re-emphasis circuit. The idea being that the user could put their CD player into the EQ circuit, then into the phono stage and see if they preferred the 'colorations' of the phono stage over the 'flat' CD input. The punchline of the joke was that the two RIAA circuits did not sum to flat response. So which one was at fault, the emphasis or de-emphasis curve? Or were they both wrong? (If memory serves, the phono stage was fairly close but the little CD-emphasis dingus was way off. Draw your own conclusions.)
Measure the phono response directly. Spreadsheets are freely available on the web that show proper RIAA response down to zillionths of a dB. Start at 20Hz and watch the response fall. If it deviates from the proper response, you'll be able to tell.
And you won't have to worry about all this silly loading stuff...
Or, as people have said repeatedly...do your homework, please.
|23rd July 2008, 02:21 PM||#32|
Join Date: Apr 2006
Location: italia - ora USA -WI
well thank you for your answer.
Nevertheless i don't fully agree with you.
Just becase you don't like/use simulators doesn't mean that they are useless.
In fact i have learned by simulation that a small variation of a specific capacitor incides on the middle band and specifically on lack of middle band thus distortion.
Obviously i have a proven circuit as you said and that's a given.
But knowing how the things varies by variating some of the paramenters i think it is important/interesting as well.
This brought to my attention the fact that capacitors need to be accurately matched and of selected values.
Unfortunately we are not in school anymore and asking something doesn't necessary mean to not do homeworks as well, but thanks anyways for the good point you brought up.
|23rd July 2008, 10:33 PM||#33|
Join Date: Feb 2001
Location: Columbia, SC
Actually, if you have to have a simulator show you that changing the value of a capacitor in a low pass filter changes the response of the filter it's a pretty clear demonstration that you haven't thought things through. Simulations are not a substitute for thinking--though many treat them as such.
Randomly changing component values in a filter--any filter--always changes the response. You can change the values, but only as long as you maintain the same time constants. This isn't something I'd suggest until you've pondered the what and why of a filter for a while. Those values were not simply plucked out of thin air.
So, no, I don't think you learned anything by simulating the circuit.
The RIAA filter exists because it cuts down on the low end transit required by the cutter head. This reduces the power needed when cutting the master, allows for more music to be placed on an album side, and makes it easier for the playback cartridge to track the music. It's not a win-win proposition, however. By boosting the low end on playback, it increases rumble and woofer excursion caused by warped records.
It would be nice if a filter wasn't needed, but there's not a practical way to record (or play back) albums without it, even now. In fact, there's not a signal medium (short of playing music live, and even that tends to involve filters of one sort or another, though it doesn't have to) that doesn't require a filter of some sort:
--Records: RIAA filter. (at least two variants exist, and historically there were other filters before the RIAA)
--Tape: Any of several filters, depending on the era, tape heads, etc. (You want to get kinky? Go back to wire recorders.)
--Radio: It doesn't even bear thinking about. There are filters of one sort or another all through tuners, not to mention a 19kHz filter in some models.
--CD: The infamous "brick wall" filter at 22kHz...'nuff said...
And so on...
Tinker with the values of the filters at your peril. They're there for a reason. You shouldn't need someone or something (e.g. a simulation) to point out the obvious: Changing things changes things.
|Thread Tools||Search this Thread|
|Thread||Thread Starter||Forum||Replies||Last Post|
|Some Quad els57 questions! - Novice Questions!||Fanuc||Planars & Exotics||13||5th December 2007 10:26 AM|
|PSU for QSC amp questions||rkc7||Parts||5||10th November 2005 06:18 PM|
|Last questions about this||Madmike2||Multi-Way||9||14th June 2005 12:12 AM|
|Questions, questions, from newb DIYer||Phatman91624||Multi-Way||41||3rd August 2002 03:52 AM|
|New To Site?||Need Help?|