I guess if one is getting the kinds of overload resulting from dirt clods and potholes, it's time to clean or just discard the disk!
The OL margins that are brought out in JA measurement sidebars are based on nominal 5mV rms at 1kHz and 50mV rms at 20kHz, although he periodically mentions that there's not likely too many records with even that level at 20kHz. But then he uses that number to determine the overload margin for the unit under scrutiny, so that's been what I've figured on.
In the case of the preamp presented, the first stage gain is (4.7k/100) +1, i.e. times 48, +33.6dB. The output swing typical of an OP37 according to TI with +/-15V rails is +/- 13.4V. So the maximum peak input voltage below clipping to that stage is 279.2mV, or for a sinusoid, 197.4mV rms. So the OL margin based on the above 50mV rms is a factor of 3.948, or 11.9dB, of course improving at lower frequencies.
Now this may be plenty in almost every case, but if you are submitting a product for review and want accolades in the measurement sidebars, this wouldn't cut it.
The voltage noise density of the OP37 is reasonably low, about 3.7nV/rt Hz at 10Hz, 3.2nV/rt Hz at 1kHz, according to TI curves. The 100 ohm divider R at about 1.3nV/rt Hz will not contribute much compared to these. The current noise density according to the datasheet is around 2pA/rt Hz at 10Hz, corners around 70Hz and reaches 400fA/rt Hz midband and above. The current noise of the 47k termination resistor is 594fA/rt Hz. Hence the total parallel noise density from those two sources will be 716fA/rt Hz midband. These parallel-noise contributions will only be somewhat important with high inductance sources at relatively high frequencies, and all of this stuff will be followed by the EQ which will proportionally reduce the HF stuff.
This is about the point in a textbook when the weary author says The calculation of the overall signal-to-noise performance is left as an exercise for the reader, and may then say See Problem 426.
Brad
Edit: these numbers for OL are actually a little optimistic, as I didn't notice the cap multiplier Q's and their ~1 Vbe drops. The OP37 does handle +/-18V btw, although that would be walking near the precipice.
Last edited:
Brad, you want to get them before they multiply.
I generally hate people who say Well, it could be worse. I worked with someone at Harman who responded thus whenever I biatched about something, and I thought, Bruce, if we were in tumbrils on our way to the guillotine, and were getting close enough to see heads being lopped off, you'd say Well, it could be worse.
But with the upstairs, It could be Worse. They could be playing loud music with lots of bass to accompany their stomping around.
The previous person was the mother of the then-manager, and I could hear her snore. A rather hefty one, and when she got out of bed I quipped that the mastodon, long believed to be extinct, in fact lives on. Got a lot of complaints from people who were drinking coffee when they read that and had it come out of their nose.
That fits exactly with my point. At 25cm/s, the cartridge will have 25mV output (assuming a typical 5cm/s output of 5mV- a V15 is slightly lower). That's well below the 200mV overload.
bcarso said:
A few more significant figures than I used, but the same answer.
That fits exactly with my point. At 25cm/s, the cartridge will have 25mV output (assuming a typical 5cm/s output of 5mV- a V15 is slightly lower). That's well below the 200mV overload.
A few more significant figures than I used, but the same answer.
But of course the overload problem is at high frequencies. Again, for most clean and well-preserved disks, it's probably not a big problem. For specsmanship and a good review --- which may or may not translate into customer respect and increased sales --- it might well be a problem.
Although, (and I've only been following them closely in recent times) about the only time I recall seeing a product get panned by JA, it was the 60k$ Vitus (?) phono pre, for which he measured pretty terrible noise. Fremer thought it was the cat's meow iirc.
Hi,
Yet at least in principle you need another 14dB above these levels... Then your overload margin is -3dB!...
And if your cartridge is "loud" it may be -6dB or even -9dB...
And if you use a MC Stepup transformer with a stepup of 30 and a "loud" MC Pickup...
You get the point.
Active EQ has no such issues.
Ciao T
PS, JA also measured no overload margin at HF for my Phono, while I measured plenty, both of us with an AP2. I never did get to the bottom of what happened there... My Phono outputs 30V at 20KHz, no sweat.
Yet at least in principle you need another 14dB above these levels... Then your overload margin is -3dB!...
And if your cartridge is "loud" it may be -6dB or even -9dB...
And if you use a MC Stepup transformer with a stepup of 30 and a "loud" MC Pickup...
You get the point.
Active EQ has no such issues.
Ciao T
PS, JA also measured no overload margin at HF for my Phono, while I measured plenty, both of us with an AP2. I never did get to the bottom of what happened there... My Phono outputs 30V at 20KHz, no sweat.
Points I also made, so clearly I agree. I designed my pre for far higher margins than were needed just because.
Even for not as clean and well-preserved, published data that I've seen doesn't show anything more than 15-20dB above nominal, and even that is a rare occurrence. Velocity is the key quantity, not frequency- JA chose a 50mV standard at 20k as a torture, not because that's actually something one routinely sees.
That sort of thing is very scary. I also asked in a thread I started (and to which you contributed I think, thanks) about measuring parallel noise, and more-or-less answered my own question. JA undoubtedly uses the on-loan SYS2722 which I suspect for RIAA conformance tests has an inverse RIAA network*, but measures noise by shorting the preamp-under-test's input, and afaik has never mentioned parallel noise as such. Similarly, for distortion measurements he uses I presume the Ap generator, which is a rather poor approximation to the impedance of a cartridge most of the time. So distortion induced at the preamp input is likely underestimated.
I wonder if Jan D. happens to remember where the 36dB OL margin "requirement" appeared? Maybe it was on Popa's website and not in LA proper.
Brad
*and how accurate is it? I know people who so completely trust Audio Precision that measurements made any other way are prima facie disbelieved. I told one friend that a DSP-based EQ box from Ashly had highish output noise, which would dominate the noise out of the loudspeakers in a prototype system--- and he flat-out didn't believe me, because it hadn't been measured with an Ap!
Of course noise meaurements are fraught with peril, but I've been making them since well before AP was even a gleam in Bruce and Richard's eyes.http://es.shure.com/idc/groups/tech_pubs/@global_managed/documents/webcontent/us_pro_v15iv_ug.pdf
page 4 has some measured points over 70 cm/s - presumably intentional music recording velocity
page 4 has some measured points over 70 cm/s - presumably intentional music recording velocity
Yet, has anyone who had that cartridge ever observed mistracking? I never did, my V15-IV was a very solid tracker. It would be interesting to know where they got those numbers for the sales brochure.
In any case, those velocities, if believable, yield at most 56mV, still safely under the limit for the B-B MM stage.
In any case, those velocities, if believable, yield at most 56mV, still safely under the limit for the B-B MM stage.
I can mistrack virtually any MC cartridge with a specific direct disc record. I did it 100's of times with a number of MC cartidges and published the results in 1978, in an IEEE paper.
It is true that it is more difficult to mistrack a Shure, but the main thing is that the Shure cartridge has a 4 pole low pass filter connected to it above 20KHz, and that keeps any mistracking from being detected as such. The problem with the Shure, is that it did not sound as good as other phono cartridges, on the whole. That is why most serious audiophiles do not use Shure phono cartridges as their reference cartridge.
It is true that it is more difficult to mistrack a Shure, but the main thing is that the Shure cartridge has a 4 pole low pass filter connected to it above 20KHz, and that keeps any mistracking from being detected as such. The problem with the Shure, is that it did not sound as good as other phono cartridges, on the whole. That is why most serious audiophiles do not use Shure phono cartridges as their reference cartridge.
Perhaps it's that they stopped making that cartridge 40 years ago. At that time, it was Gordon Holt's top choice (with a G stylus), but I guess he wasn't a serious audiophile.
I can't speak for Gordon Holt, but I did comparison tests between a Shure and an Ortofon MC, and put out the extra money for the Ortofon, for my personal system in 1965. However, for many customers, Berkeley Custom Electronics, where I worked part time, recommended the Shure M91 or M93 phono cartridges for less discriminate use.
For the record, I used the Shure M3D-N21 from 1963-1965, then switching to an Ortofon SPU, and then from 1970-1971 with a Shure M91, as a house guest in someone else's home, that I had installed for them. I then returned the Ortofon, and went on to the Supex, when it became available, then EMT, FR, etc. and never looked back, except to measure the Shure for my 1978 paper. It couldn't rise-time its way out of a paper bag, until the V15-V! '-)
For the record, I used the Shure M3D-N21 from 1963-1965, then switching to an Ortofon SPU, and then from 1970-1971 with a Shure M91, as a house guest in someone else's home, that I had installed for them. I then returned the Ortofon, and went on to the Supex, when it became available, then EMT, FR, etc. and never looked back, except to measure the Shure for my 1978 paper. It couldn't rise-time its way out of a paper bag, until the V15-V! '-)
My first Dual 1019 TT in 1970 was delivered with a Shure M75 cartridge. I kept it until 1977 (or was that 1978?), and had many opportunities to compare it with the V-15 series. These were sometimes mounted in much better TTs than mine was, being a sort of a middle-of-the-road model.
To my ears, the difference in price between a V-15 and my M75 was far greater than the difference in the sound, which was on the side of V-15. It was the better cartridge by a relatively small margin for a relatively large price difference.
Then I discovered Ortofon's LM series. I bought into it and have stayed there ever since. The only hassle I had with it was that I had to solder a 100 pF small cap to get the total capacitance to the 400 pF which Ortofon demanded. At the time, my reVox integrated amp had no adjustments, except for level, no R, no C.
So, personally, I tend to agree with John. However, many thought otherwise. In Europe at the time, a Shure V-15 in an SME 3009 arm was considered a standard for serious audiophilia, even if it was not regarded as the best ever.
What got me sold on Ortofon were two things: they were (to me) much more dynamic, more go and pazzazz, and they offered a cleaner treble.
Last edited:
Holt felt that the MCs of that time were quite tonally colored compared to the Shure. Maybe, I didn't have the piles of master tape comparisons that he did, but I was quite happy with the Shure. I wasn't thrilled with too many of the MCs of that era, there was a pervasive upper midrange dip and upper treble peak. After trying a bunch, the best one I owned was a Dynavector Diamond, which was very good, but not nearly as good as a Technics MM I replaced it with. When the Technics became unobtainable, I switched to a Troika, which was also pretty good, though a step down from the Technics. I'm back to an MM, though (an A-T, really quiet, good tracking, and tonally similar to digital, which I consider a virtue).
One of those fun things I learned from a major studio monitor manufacturer was that most of the first studio monitors (Think Altec to JBL dominant time) rolled off the high end a bit. That was because they were using 1" measurement microphones and did not allow for the HF rise caused by such a large diaphragm. As a result most of the records were mixed with a boosted HF content. The home reproduction systems were of course voiced to sound best with the best source of music.. records.
It took the studio monitor guys a while to fix the problem. When they showed truly flat monitor loudspeakers to the studio guys the response was they were too bright. So they basically improved about a db per decade until folks would accept flat.
So when the Shure V15 was popular a slight HF rolloff (3 db / 20K) would have sounded right.
So cartridge choice would have been influenced by the record source!
It took the studio monitor guys a while to fix the problem. When they showed truly flat monitor loudspeakers to the studio guys the response was they were too bright. So they basically improved about a db per decade until folks would accept flat.
So when the Shure V15 was popular a slight HF rolloff (3 db / 20K) would have sounded right.
So cartridge choice would have been influenced by the record source!
Now I know why I never liked the V15. I was a Grace F9e fan. I would take the Grace arm over the SME any day. Linn if you were patient enough to get it set up. I miss the cartridge wars. They were almost like religion.
Moved from SIMetrix to LTSpice. A lot more tedious data entry, but a lot more capability.
Moved from SIMetrix to LTSpice. A lot more tedious data entry, but a lot more capability.
- Status
- Not open for further replies.