Hi Greg,
My first IC was to buffer/match the cartridge and provide frequency invariant source impedance for the 1/2 RIAA IC.
So that the 1/2 RIAA IC would work accurately, it was then not load burdened by the 1/2 RIAA passive filter.
This too was then buffered by a final IC which drove output transistors for 50 ohm line level output with good headroom. ( I had 20ft coax/BNC to LS sited monoblocks.)
A single IC would have had source impedance and noise level varying with frequency at differential input, and line driver loading could affect RIAA feedback accuracy/capability.
There was not any extra noise or distortion, and when there are not any commercial limitations you can build better. The 1/2 RIAA IC followed by a 1/2 RIAA passive/ground connected filter was actually quieter, as well as cleaner/clearer sounding, than full direct differential RIAA on a single IC. Additional noise after the passive filter was negligible.
Modern ICs are now so much better, and I note there are several commercially available single IC based 'phono boxes' costing up to circa £150. Many makers also go on about component selection and tolerance, but that does not make them the ones that get top flight reviews.
Hi Andrew,
I think your reply covers my point about it 20/50Hz roll-off being optional.
Some others clearly think it worthwhile offering that facility too.
Yes 20Hz could kill music, and a brick wall mounted deck becomes the minimum essential, but not everyone can afford to buy or house Tannoys etc, and thus the 20Hz option can have its place.
Yes Bear, I was happy with it. Silly me for selling it !
Cheers ......... Graham.
My first IC was to buffer/match the cartridge and provide frequency invariant source impedance for the 1/2 RIAA IC.
So that the 1/2 RIAA IC would work accurately, it was then not load burdened by the 1/2 RIAA passive filter.
This too was then buffered by a final IC which drove output transistors for 50 ohm line level output with good headroom. ( I had 20ft coax/BNC to LS sited monoblocks.)
A single IC would have had source impedance and noise level varying with frequency at differential input, and line driver loading could affect RIAA feedback accuracy/capability.
There was not any extra noise or distortion, and when there are not any commercial limitations you can build better. The 1/2 RIAA IC followed by a 1/2 RIAA passive/ground connected filter was actually quieter, as well as cleaner/clearer sounding, than full direct differential RIAA on a single IC. Additional noise after the passive filter was negligible.
Modern ICs are now so much better, and I note there are several commercially available single IC based 'phono boxes' costing up to circa £150. Many makers also go on about component selection and tolerance, but that does not make them the ones that get top flight reviews.
Hi Andrew,
I think your reply covers my point about it 20/50Hz roll-off being optional.
Some others clearly think it worthwhile offering that facility too.
Yes 20Hz could kill music, and a brick wall mounted deck becomes the minimum essential, but not everyone can afford to buy or house Tannoys etc, and thus the 20Hz option can have its place.
Yes Bear, I was happy with it. Silly me for selling it !
Cheers ......... Graham.
Hi Bear and Graham,
When I designed the Eidetic GB1a preamp back in 1989, I used the newly released LT1028, at odds with my past designs which used discrete (2SB737 and/or 2SD786 from Rohm and various discrete and integrated topologies).
I tested it thouroughly for input impedance interaction which was I believed to be virtually solely the domain of single ended input stages. It is virtually eliminated by diff'l inputs isolating the input from the feedback and the large amount of loop gain available for feedback from such a chip. So the input Z was, to all intents and purposes, identical to the defining components R,C on the inputs.
In my case this stage was switchable for 3 R (100K,47K,100)loadings plus a user define addition (solder pegs), 3 C (100p,200p,320p) loadings plus a user define addition(solder pegs), and three loop gain settings (34dB,44dB,54dB).
A diff'l input will always degrade the noise performance by 3 dB over a single ended stage, but it was still some 20dB lower than record surface noise so considered acadaemic, still achieving 92dB weighted by Australian Hi Fi on test (whole pre/power input to output) and specified at 76dBA (re 1mV) MC and 84dBA (re 5mV) MM.
Australian Hi Fi singled out the phono preamp stage performance as -
"The RIAA response is 0.4dB high in the bass region then tightens up to be +/- 0.2 dB across the rest of the audio band.....the overload margin was an excellent 37.84 dB - one of the best results we have seen for quite a while"
In summary input impedance interaction, the discovery of which spawned the buffer and multiple stage approach is easily eliminated through diff'l input and modern low noise monolithic chip technology.
Greg
When I designed the Eidetic GB1a preamp back in 1989, I used the newly released LT1028, at odds with my past designs which used discrete (2SB737 and/or 2SD786 from Rohm and various discrete and integrated topologies).
I tested it thouroughly for input impedance interaction which was I believed to be virtually solely the domain of single ended input stages. It is virtually eliminated by diff'l inputs isolating the input from the feedback and the large amount of loop gain available for feedback from such a chip. So the input Z was, to all intents and purposes, identical to the defining components R,C on the inputs.
In my case this stage was switchable for 3 R (100K,47K,100)loadings plus a user define addition (solder pegs), 3 C (100p,200p,320p) loadings plus a user define addition(solder pegs), and three loop gain settings (34dB,44dB,54dB).
A diff'l input will always degrade the noise performance by 3 dB over a single ended stage, but it was still some 20dB lower than record surface noise so considered acadaemic, still achieving 92dB weighted by Australian Hi Fi on test (whole pre/power input to output) and specified at 76dBA (re 1mV) MC and 84dBA (re 5mV) MM.
Australian Hi Fi singled out the phono preamp stage performance as -
"The RIAA response is 0.4dB high in the bass region then tightens up to be +/- 0.2 dB across the rest of the audio band.....the overload margin was an excellent 37.84 dB - one of the best results we have seen for quite a while"
In summary input impedance interaction, the discovery of which spawned the buffer and multiple stage approach is easily eliminated through diff'l input and modern low noise monolithic chip technology.
Greg
andy_c said:
I don't know about the SX74/VG74, but on the schematics I saw
of the older VG66 cutting rack there was a first-order at 50kHz,
in addition to a second order low pass at about 30kHz. And that is then without the treble limiter. All of these there to protect the cutting coils.
I conclude from this that above 10kHz there is no such thing as a RIAA standard.
I acted by modifying a cheap Rotel phonostage I had lying around for a variable 75us de-equalisation, offering 5 steps of each 1dB boost at 20kHz. During replay I can click through them and listening to voices it is easy to find the correct replay curve.
amplifierguru made an excellent point.
There are many factors that could potentially have an impact on an electronic device. Some of them are major, some of them are minor. It is the responsibilities of a good engineer to figure out what to prioritize and focus only on those that are important.
Otherwise, we will end of chasing down things that are sound sophisticated in theory but bear nothing in reality.
There are many factors that could potentially have an impact on an electronic device. Some of them are major, some of them are minor. It is the responsibilities of a good engineer to figure out what to prioritize and focus only on those that are important.
Otherwise, we will end of chasing down things that are sound sophisticated in theory but bear nothing in reality.
ampguru,
I'm really not questioning your product's specs or results... and quite honestly, I beg forgiveness for not really looking at either LTs or AD's chip offerings much for some time now... so I'd have to go and pull the book or CD and see this chippie! Sounds nice as you describe it.
My point is simpler than this. It is that there may be "hidden" sonic nirvana in the "split passive" approach to RIAA. Indeed in any given systems any "improvement" may be moot (?) but the experiences I've had purely on the listening end indicate that in the same way there are (oddly) audible diffs to be percieved in "ultra low distortion" power amps and with solid state vs. tubes, a similar set of things happens in LP reproduction.
The first point I made is that since LP is hardly a "technically accurate" medium, that may be justification for simply "making it sound good."
For the solid state implementations, the best I've heard so far are split passive FET/Mosfet designs, but they rarely are as nice to listen to as are split passive FET/tube RIAA implementations! In each case the deviation from the RIAA curve is not likely to be a major contributor to the percieved quality. And in any event, there wasn't much in any implementation I refer back to.
Hope this clarifies?
_-_-bear
I'm really not questioning your product's specs or results... and quite honestly, I beg forgiveness for not really looking at either LTs or AD's chip offerings much for some time now... so I'd have to go and pull the book or CD and see this chippie! Sounds nice as you describe it.
My point is simpler than this. It is that there may be "hidden" sonic nirvana in the "split passive" approach to RIAA. Indeed in any given systems any "improvement" may be moot (?) but the experiences I've had purely on the listening end indicate that in the same way there are (oddly) audible diffs to be percieved in "ultra low distortion" power amps and with solid state vs. tubes, a similar set of things happens in LP reproduction.
The first point I made is that since LP is hardly a "technically accurate" medium, that may be justification for simply "making it sound good."
For the solid state implementations, the best I've heard so far are split passive FET/Mosfet designs, but they rarely are as nice to listen to as are split passive FET/tube RIAA implementations! In each case the deviation from the RIAA curve is not likely to be a major contributor to the percieved quality. And in any event, there wasn't much in any implementation I refer back to.
Hope this clarifies?
_-_-bear
Hi tlf,
Yes indeed. And of course individuals do prioritise their requirements from different viewpoints, thus individual ideals equals diversity of choice + experience + learning, and this irrespectively of whether the great minds that are said to think alike - agree or not.
Hi Greg,
I take the view that NFB derived control can sound correct *only* when the (NFB loop controlled) node that is generating the output is not adversely challenged by any additional components, devices, interconnects, the often non-resistive and sometimes reactive load, or the feedback loop itself; and thus the resultant dynamic quality of the circuit's reproduced sound cannot be determined by measuring with steady sinewaves or through simulated parameters alone.
Cable interconnects are capacitive, and a subsequent IC or decent follower stage provides low impedance output whilst simultaneously isolating possible inter-reaction from an RIAA stage, as does a sensibly arranged passive 1/2 RIAA filter with its own buffer.
There is much more to appreciate than technological 'ideals'. Indeed there remains a need to build, test and real-world compare, which thankfully diyAudio members do on a very broad scale.
I did not know that there had been a discovery of input impedance interaction and a spawning of the buffer plus multiple stage approach, for I had constructed and tested in isolation, but I cannot take for granted your statement that this concern has been eliminated via the use of modern ICs.
Composite circuits can audibly alter phase linearity or generate an unexpected dynamic response in a manner that cannot be shown via amplitude measurement.
Hi Werner,
Yes, and different labels / productions reproduce differently at higher frequencies, as do different cartridges, such that only one pre-settable characteristic might sound more correct with a chosen combination.
Such variations are very rarely catered for, thus in reality listeners can still lose out with a fixed RIAA phono stage, no matter how accurate it is.
This was my reason for having a twin gang variable capacitor and twin gang potentiometer loading the cartridge as it fed the input buffer, for the resultant adjustable reactive loading could additionally lift or cut / loosen or dampen the upper registers to optimise the response in a manner that no amount of subsequent equalisation could ever do.
Hi Bear,
It is good to read your comments here.
I note you suggest that the best cartridge connection is via the lightly capacitive FET.
Is this a follower or differential input where the minute current flowing is out of phase with the voltage generated, and thus with input current a minimum when resistive (or possible bipolar input) cartridge winding loading is at a maximum. ?
I could imagine a slightly different reproduction quality being the comparative result.
Cheers ......... Graham.
Yes indeed. And of course individuals do prioritise their requirements from different viewpoints, thus individual ideals equals diversity of choice + experience + learning, and this irrespectively of whether the great minds that are said to think alike - agree or not.
Hi Greg,
I take the view that NFB derived control can sound correct *only* when the (NFB loop controlled) node that is generating the output is not adversely challenged by any additional components, devices, interconnects, the often non-resistive and sometimes reactive load, or the feedback loop itself; and thus the resultant dynamic quality of the circuit's reproduced sound cannot be determined by measuring with steady sinewaves or through simulated parameters alone.
Cable interconnects are capacitive, and a subsequent IC or decent follower stage provides low impedance output whilst simultaneously isolating possible inter-reaction from an RIAA stage, as does a sensibly arranged passive 1/2 RIAA filter with its own buffer.
There is much more to appreciate than technological 'ideals'. Indeed there remains a need to build, test and real-world compare, which thankfully diyAudio members do on a very broad scale.
I did not know that there had been a discovery of input impedance interaction and a spawning of the buffer plus multiple stage approach, for I had constructed and tested in isolation, but I cannot take for granted your statement that this concern has been eliminated via the use of modern ICs.
Composite circuits can audibly alter phase linearity or generate an unexpected dynamic response in a manner that cannot be shown via amplitude measurement.
Hi Werner,
Yes, and different labels / productions reproduce differently at higher frequencies, as do different cartridges, such that only one pre-settable characteristic might sound more correct with a chosen combination.
Such variations are very rarely catered for, thus in reality listeners can still lose out with a fixed RIAA phono stage, no matter how accurate it is.
This was my reason for having a twin gang variable capacitor and twin gang potentiometer loading the cartridge as it fed the input buffer, for the resultant adjustable reactive loading could additionally lift or cut / loosen or dampen the upper registers to optimise the response in a manner that no amount of subsequent equalisation could ever do.
Hi Bear,
It is good to read your comments here.
I note you suggest that the best cartridge connection is via the lightly capacitive FET.
Is this a follower or differential input where the minute current flowing is out of phase with the voltage generated, and thus with input current a minimum when resistive (or possible bipolar input) cartridge winding loading is at a maximum. ?
I could imagine a slightly different reproduction quality being the comparative result.
Cheers ......... Graham.
Hi andy C,
Interesting your info re the Neumann 50KHz and your website RIAA derivation - particularly as my design I spoke of earlier almost replicates this - It was wrapped around an LT1028 with 34dB 1KHz gain (MM setting) and an added 2K2 to the figure of 8 RIAA network with approx 1K feedback R to ground. I was, in fact creating a zero up there around 50KHz and actually let the accumulated rolloffs through the chain (my power amplifier had input and feedback 160KHz filters) so in essence I had restored most of the Neumann TC. For what it's worth I ran the low bass time constant down to 15Hz hence the 0.4dB rise ( I was getting mixed messages about the IEC rolloff so met it halfway) and midband was measured accurate to 0.02dB.
I can recall (30years) the process I went through with my little Sugden manual test set I imported from the UK. I built an accurate reverse RIAA network to feed the sig gen into so I could test each preamp by sweeping the ranges and waiting for the thermistor to stop bouncing.
Re Cheech and Chong, tried Shel Silverstein?
Hi Bear, agree with you on the input fets - I was meaning s.e. bipolar inputs.
Cheers,
Greg
Interesting your info re the Neumann 50KHz and your website RIAA derivation - particularly as my design I spoke of earlier almost replicates this - It was wrapped around an LT1028 with 34dB 1KHz gain (MM setting) and an added 2K2 to the figure of 8 RIAA network with approx 1K feedback R to ground. I was, in fact creating a zero up there around 50KHz and actually let the accumulated rolloffs through the chain (my power amplifier had input and feedback 160KHz filters) so in essence I had restored most of the Neumann TC. For what it's worth I ran the low bass time constant down to 15Hz hence the 0.4dB rise ( I was getting mixed messages about the IEC rolloff so met it halfway) and midband was measured accurate to 0.02dB.
I can recall (30years) the process I went through with my little Sugden manual test set I imported from the UK. I built an accurate reverse RIAA network to feed the sig gen into so I could test each preamp by sweeping the ranges and waiting for the thermistor to stop bouncing.
Re Cheech and Chong, tried Shel Silverstein?
Hi Bear, agree with you on the input fets - I was meaning s.e. bipolar inputs.
Cheers,
Greg
Hi Greg,
Glad you enjoyed the web page and it's interesting to hear your experiences on this. I remember doing square wave testing of phono preamps years ago with an inverse RIAA network. There was a bunch of overshoot in the square wave response. Turns out the unspecified f4 frequency (a pole for the inverse network and a zero for the phono stage) didn't match up between the phono amp and the inverse network. So the square wave response was like that of an RC network having a zero at a frequency f4a, followed by a pole at a higher frequency f4b. In this case, f4a was the zero frequency of the phono preamp (which may have been something like 75 kHz or so) and f4b was the pole of the inverse RIAA network (which I think was about 300 kHz or so, as it was a copy of the Lipshitz network).
Either that, or it was some kind of discontinuity in the time-space continuum, interacting in an undesirable way with my random audio BS generator.
Regarding Shel Silverstein, I did hear one song he wrote called The Perfect High from an LP by Bird and McDonald called The Dukes of Disgusting. This got a lot of play on the Dr. Demento show back then. It might be worth getting the vinyl rig working again just to hear that one. I don't think it's available on CD.
I'll have to bug Werner for more info, as it looks like my web page should be corrected to show that a single time constant isn't enough to compensate. Or maybe it depends on the model of the cutting lathe? Drat.
Glad you enjoyed the web page and it's interesting to hear your experiences on this. I remember doing square wave testing of phono preamps years ago with an inverse RIAA network. There was a bunch of overshoot in the square wave response. Turns out the unspecified f4 frequency (a pole for the inverse network and a zero for the phono stage) didn't match up between the phono amp and the inverse network. So the square wave response was like that of an RC network having a zero at a frequency f4a, followed by a pole at a higher frequency f4b. In this case, f4a was the zero frequency of the phono preamp (which may have been something like 75 kHz or so) and f4b was the pole of the inverse RIAA network (which I think was about 300 kHz or so, as it was a copy of the Lipshitz network).
Either that, or it was some kind of discontinuity in the time-space continuum, interacting in an undesirable way with my random audio BS generator.
Regarding Shel Silverstein, I did hear one song he wrote called The Perfect High from an LP by Bird and McDonald called The Dukes of Disgusting. This got a lot of play on the Dr. Demento show back then. It might be worth getting the vinyl rig working again just to hear that one. I don't think it's available on CD.
I'll have to bug Werner for more info, as it looks like my web page should be corrected to show that a single time constant isn't enough to compensate. Or maybe it depends on the model of the cutting lathe? Drat.
Yes similar experience with the sq wave testing - I simply filtered it above 100KHz - didn't have much overshoot then! I still do that in my amps - an input filter works wonders to limit ringing on C loads as the UHF harmonics simply aren't stimulated and stability is enhanced. After optimisation of course.
Re Shel Silverstein, he's a writer of children's poetry,etc.. but his 'adult' offerings have titles like 'Someone ate the baby' and '(Everybody say's I'm) Paranoid' Searching either of those should turn up some gems.
Cheers,
Greg
Re Shel Silverstein, he's a writer of children's poetry,etc.. but his 'adult' offerings have titles like 'Someone ate the baby' and '(Everybody say's I'm) Paranoid' Searching either of those should turn up some gems.
Cheers,
Greg
I will look for more of his stuff. Even though it's been years since I've heard the song, I still remember these words:
"...old men or bright-eyed youth,
It's always easier to sell them some **** than it is to tell them the truth."
Classic. We should put him to work selling microwave tweeters
.
"...old men or bright-eyed youth,
It's always easier to sell them some **** than it is to tell them the truth."
Classic. We should put him to work selling microwave tweeters
.lt cdr data said:
The first circuit is by a man named Graham. Mind, you have a number of R values wrong, and where is the input bias circuit?
The second circuit is by a man named Tom. I think you have a couple of passive, maybe even active, components missing. U2/3 is likely an NE5532, and the 10n and 150n caps are ceramics.
--
Andy,
it is pointless to try to make a single-fits-all RIAA stage. A VG74 cutting rack probably differs from a VG66 in the treble, and then we don't even know what Ortofon, Westrex, ... did.
And even if we did, and even when all these cutting amps had the very same treble characteristics, we are still confronted with the gradually dropping treble contents of each successive record as the stamper wears out.
RIAA CIRCUIT = 12V supply with NE5532
Hi DIY FriendS,!!
I have ,many problems to search a RIAA circuit to put in my power Amplifier.
Do you have any circuit (12V battery supply and low cost components) to input in my PCB layout?!
I have found others SCH with NE5532... Do you help me?!
Schematic without problems with ignition,alternator,motor (external boops...)
my configuration:
INPUT=RCA PLUGS
OUTPUT=CIRCUIT AMPLIFIER WITH TRANSISTORS.,,
Please, help....mee!
Thanks soo lot.. Best Regards...
sory my english is very poor
Caio Fr.
caio.sul@smar.com.br
caio@digisound.com.br
Hi DIY FriendS,!!
I have ,many problems to search a RIAA circuit to put in my power Amplifier.
Do you have any circuit (12V battery supply and low cost components) to input in my PCB layout?!
I have found others SCH with NE5532... Do you help me?!
Schematic without problems with ignition,alternator,motor (external boops...)
my configuration:
INPUT=RCA PLUGS
OUTPUT=CIRCUIT AMPLIFIER WITH TRANSISTORS.,,
Please, help....mee!
Thanks soo lot.. Best Regards...
sory my english is very poor
Caio Fr.
caio.sul@smar.com.br
caio@digisound.com.br
Hi Mauro,
I am not in a position to make any real-world reverse testing measurements or specrutm anslysis.
I feel that reverse (back EMF) injection can cause a NFB loop induced problem that will modulate forward amplified signal, and that this will show on a spectrum analyser with say 10kHz creating additional interference to say a 1kHz signal input; similar to your dumping factor test set-up in Post#70.
I assume that the other 'Results (preliminary)' set-up was to check for internal damping response phase shift induced error.
Different word usage between our native languages prevents me from fully understanding your last posts, and I did not realise that you were waiting for comment. I was waiting to see if you had any findings from those tests.
I wish you luck in completing those tests.
Cheers ......... Graham.
I am not in a position to make any real-world reverse testing measurements or specrutm anslysis.
I feel that reverse (back EMF) injection can cause a NFB loop induced problem that will modulate forward amplified signal, and that this will show on a spectrum analyser with say 10kHz creating additional interference to say a 1kHz signal input; similar to your dumping factor test set-up in Post#70.
I assume that the other 'Results (preliminary)' set-up was to check for internal damping response phase shift induced error.
Different word usage between our native languages prevents me from fully understanding your last posts, and I did not realise that you were waiting for comment. I was waiting to see if you had any findings from those tests.
I wish you luck in completing those tests.
Cheers ......... Graham.
Interesting NP.
As I see it, the typical op amp response is near ideal for single loop feedback RIAA MC/MM preamps. The LT 1028 is a near perfect embodiment with 10Hz ol bandwidth and 6dB per octave decline in gain from 140dB. The RIAA equalisation varies by only about 6 dB from this over the audio 20 - 20KHz spectrum. So the feedback only varies by this much - for those who think it matters!
Just think a near constant feedback factor from 20 -20KHz from an op amp!! And constant feedback factor means constant Zin (support).
There's a valid argument for an integrator input stage followed by a 500Hz -2K12Hz +12dB inflection as a RIAA concept (plus 20Hz and 50KHz ends of course).
Greg
As I see it, the typical op amp response is near ideal for single loop feedback RIAA MC/MM preamps. The LT 1028 is a near perfect embodiment with 10Hz ol bandwidth and 6dB per octave decline in gain from 140dB. The RIAA equalisation varies by only about 6 dB from this over the audio 20 - 20KHz spectrum. So the feedback only varies by this much - for those who think it matters!
Just think a near constant feedback factor from 20 -20KHz from an op amp!! And constant feedback factor means constant Zin (support).
There's a valid argument for an integrator input stage followed by a 500Hz -2K12Hz +12dB inflection as a RIAA concept (plus 20Hz and 50KHz ends of course).
Greg
Nelson Pass said:
Mr.Pass,
could you comment on the full signal path RIAA accuracy of that
cutting setup, before and after mods? You may have seen my remarks
on the Neumann VG66 above, but I am very interested in learning
details of other 'Verstaerkungsgestelle' (love that term, makes
it sound like a mad-scientist weapon).
Graham Maynard said:
Graham,
Sorry, been away from the black hole of the internet for a few days...
Obviously, the FET has much to commend it for connecting to those pesky phono cartridges!
Usually the aim is to get low noise and copious amounts of gain for MC carts, some of which have scary little output! The resulting design lately has been a cascode with a nice tube (one that has gold plated grids and in the triode variety) on top for the first stage. That's resulted in the most glorious "golden tone"!
Not sure, but I may have been the first to use a FET/Tube cascode for the front end of a phono stage... but probably not, after all just about everything was done somewhere earlier.
As I mentioned, for LP reproduction it seems reasonable to err ever so slightly over to the side of just a touch of "euphonia", so imho differential inputs or stages are to be avoided to this end... even so the results are technically surprisingly good imho. And done right, it sure sounds fabulous...
_-_-bear
amplifierguru said:
That was my thinking when I used it in the Forte 2 preamp.
Werner said:
The Ortofon cutting head has a large midband mechanical
resonance which has to be flattened with feedback. As I
indicated, the original system used a large number of op amps
(of late 60's vintage) in the signal path, some doing RIAA
reverse equalization, some providing gain, some processing
feedback and some adding additional equalization for the
cutterhead response. The actual response of this system
was quite flat, but MoFi judged that the sonics of the system
suffered from the complexity of the signal path through old
op amps and a power amplifier which looked like a variation
of a Phase Linear 700.
What I did was high accuracy measurement of the response
of the existing system and I set out to duplicate it with a much
simpler circuit. I developed a very accurate model of the
cutterhead's mechanical and electrical characteristic and began
a series of computer simulations. I concluded that the whole
thing could be duplicated with networks around one high quality
gain stage, but it was very important that the circuit have a
large open loop gain. It was a lot easier to choose an LT1028
than to try to build such a high open loop device from discrete
parts. This circuit then drove the cutterhead amp which was
a very simple complementary Class A mosfet circuit.
In the end I was able to duplicate the response of the original
within a fraction of a decibel, but with a lot fewer "electronic"
artifacts. After some tweaking by MoFi, it was judged good,
and they mastered the last few years vinyl on it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.