|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
173.160.222.138
In Reply to: RE: This is patently false for 2 reasons: posted by Ralph on May 31, 2017 at 09:02:47
"no need for matched resistors to simulate that"
Wouldn't you need that center point voltage to use as a reference for things like RIAA filter reference if, for example, you were doing a full active riaa stage?
I just meant to point out that even with the coils true midpoint buried in the middle of the cartridge coil it isn't impossible to know the voltage, or at least come very close.
Follow Ups:
is what you are looking for.
In the case of the RIAA network, it can be done differentially so you don't need dual matched resistors there either (although you do want them to be correct for the right curve of course).
For example in our preamp the EQ network calls for dual 110K resistors in one location but they can be replaced by a single 220K resistor. In this way if one half of the balanced circuit does not perform as well as the other half, the EQ is unaffected.
The prior example is also a bit of myth-busting about the idea that you need twice as many parts as single ended; obviously you don't; our EQ network has the same number of parts as a single-ended embodiment, despite being balanced differential.
I came up with another balanced circuit puzzle.A front end circuit I'm playing with is using a servo in order to eliminate ultra low frequency so I can DC couple it.
Is there a better way to derive a servo reference point than the two matched resistor method?
Edits: 06/01/17
What the ' two matched resistor method ' is. Do you have a schematic?
As a tip, if using a servo, I think you will find that 2 poles in the servo's operation will work a lot better than one!
My servo attempts to bring the preamps output offset to the same potential as the midpoint of the cartridge coil by stripping out the low frequency from the output and (negative) feeding it back.
All I meant by "two matched resistor method" is by placing two matched resistors in series across the coil outputs, the midpoint between the resistors may be usable as the servo reference.
Unless there is some better way to find the cartridge midpoint than using two "matched" resistors which will never actually be matched in the real world....I'd love to find a better way but have not come up with anything.
Sorry that was unclear.
I'm at work now. I can hack up a schematic when I get home and post it.
My servo circuit uses more than two poles. It is based on a weird circuit I found on the internet in an old LT app note and mixed in a bit of my own special sauce but all seems to simulate nicely. I get very deep audio band rejection but takes maybe a bit too long to settle. I will need to fine tune it when I get hardware in hand, ie this is all simulation at this point.
Just use ground as a reference for the servo, so the DC output is 0V.
As far as the input, each preamp side will have its own resistor to ground so that the preamp can function; that value should be 23.5K so that the load on the cartridge is 47K.
Thank you for your replies. They are very much appreciated. I had some ideas about things to try but it sure instills confidence to talk to someone who has tried some of this stuff.The plan all along has been to make it easy to try either ground referenced or floating referenced servos via jumper setting.
My desire to leave the servos referenced to a floating , non ground node was triggered by these fully balanced/differential op amps, apparently targeted at the audio market, such as the OPA1632.
These parts have integrated servos to set output common mode voltage. The app notes suggest common usage being tying the Vocm pin to the ADC reference. To my mind this implies there are must be audio market ADC's out there using external non ground voltage references.
My concern is that if I wanted to use one of these ADC's and I had my front end servos referenced to ground I may exceed the common mode range of my of my output stage under certain large signal swing circumstances.
My thinking was that floating front end servo references or tying it to my own version of a Vocm reference pin and an output stage that allows setting output common mode voltage (similar to OPA1632 et al) gets me around that problem.
Edits: 06/03/17 06/03/17 06/03/17 06/03/17
-
Well, building may be putting it just a bit strong. I just sit around using gigawatt/hours simulating then changing my mind and then do that some more.
I think I might be getting close...but I've said that before and then it evolves.
At the moment I'm fairly excited about some breakthroughs I'm having on the bass processing circuitry I'd like to include.
It's always something.
Who knows, maybe someday I'll start laying out a PCB. haha
Zoooo.. a daydreamer speculating on what if.. and a man with a horse in the race .. hoping to make 10s of $.
Not the most enlightening of threads.
And yet you read to the end.....lol
I would guess the general life misery level must be pretty high for someone as apparently clueless to feel the need to comment in such a worthless manner.
I hope things turn around for you soon.
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: