But Richards current mode design is so simple you can build several each optimised for a different cartridge you own. For DIY fun it has a lot of intrigue (as do the other 6 phono projects currently in line
)Guru Wurcer, I think your field recorder example isn't representative of many live sound scenarios.
For those unversed in the art, the overload margin required is dependent on the usual operating position of your volume control or fader.
If your fader is set to -10dB (usually marked '0dB') from max (usually marked '10dB'), you only need 10dB of overload. But if you habitually operate with it at '-20dB', you need 30dB overload margin else there will be times when the input stage will overload before the output in a live sound/recording situation.
With a domestic volume control, the equivalent positions are 12 o'clock (needing 20dB margin) and 0930 (usually needing 30dB margin)
In da 21st century, with all digital sources and their nominal 2V o/p, these requirements are much relaxed as you know EXACTLY what to expect in modern domestic stuff.
But I agree with Bonsai about 30dB for domestic vinyl or for live sound / recording. From the early 80's, 30dB overload was easy (& usual) with ju.ust over 20dBu max from 5532s and operating at -10dBu.
But its nice to have your volume control over 1200 as that is usually a better matched section of the stereo pot. Yus kids wid da digital volume controls (inventions of da devil
) pls excuse an old man reminisciing about potentiometers.Common base i/p like my amp, or better still a virtual earth, on your MC amp helps even out the different MC cartridge sensitivities. You can check this out with the list of MC cartridges in my Yahoo MicBuilders document.
Last edited:
Useless with a good design, a real pre-amp has no recovery time. A pre-amp also has a gain pot. We just disagree a fixed gain pre-amp in front of a field recorder with a gain adjust works just fine for LP transfer. The Pure Vinyl folks have LOTS of happy customers. I've tried several hot recordings including the Sheffield track record, no problem.
Scott, the Tomlinson Holman article is the go to reference on this issue.
Not discounting what you are saying, but his paper drew on multiple sources plus his own experiments to plot the outer envelope of overload requirements. Admittedly, one or two of the recordings that were unearthed were truly astounding (one of them ~80cm/sec peak O/P) but the general consensus from what I can glean from multiple sources is that at the minimum you should cater for 25cm/sec (~14dB ref 5cm/sec) + then the other stuff I mentioned earlier.
Re the crackels and pops - we are probably misunderstanding each other. I have not said these cause 'overload recovery time', rather the issue is if they cause the amplifier to clip. I have a Chick Corea vinyl and can play it at a quiet level (10-11 on the volume dial) but there is a scratch or pock mark on part of the track for 1 or 2 revolutions that must be at least 10 or 15 dB above the music level. I'd hate to think how a passive RIAA with all the gain up front would handle that - at least with an all active solution the HF gain is dropping off at 20dB/decade after 2.1 kHz . You already have loads of HF garbage and harmonics in defects like that and then to possibly clip on top of it will only make it worse and draw attention to it.
But all this aside, what specifically is your beef with a product that delivers on every spec that you as an engineer would hold dear (noise, distortion, response conformity) and also then provides 30 dB O/L margin catering for just about any recording or cart you would throw at it? Using opamps - the go to solution for MM IMV - with +-15V rails you are home and dry and cover all eventualities.
Not discounting what you are saying, but his paper drew on multiple sources plus his own experiments to plot the outer envelope of overload requirements. Admittedly, one or two of the recordings that were unearthed were truly astounding (one of them ~80cm/sec peak O/P) but the general consensus from what I can glean from multiple sources is that at the minimum you should cater for 25cm/sec (~14dB ref 5cm/sec) + then the other stuff I mentioned earlier.
Re the crackels and pops - we are probably misunderstanding each other. I have not said these cause 'overload recovery time', rather the issue is if they cause the amplifier to clip. I have a Chick Corea vinyl and can play it at a quiet level (10-11 on the volume dial) but there is a scratch or pock mark on part of the track for 1 or 2 revolutions that must be at least 10 or 15 dB above the music level. I'd hate to think how a passive RIAA with all the gain up front would handle that - at least with an all active solution the HF gain is dropping off at 20dB/decade after 2.1 kHz . You already have loads of HF garbage and harmonics in defects like that and then to possibly clip on top of it will only make it worse and draw attention to it.
But all this aside, what specifically is your beef with a product that delivers on every spec that you as an engineer would hold dear (noise, distortion, response conformity) and also then provides 30 dB O/L margin catering for just about any recording or cart you would throw at it? Using opamps - the go to solution for MM IMV - with +-15V rails you are home and dry and cover all eventualities.
As there are at least two Holman articles and a confusion over them caused you get very grumpy with me recently can you quote the title to avoid that confusion this time. The 'astounding' cut was one bit of one track on an obscure mono recording that no one has (but I will be buying soon to try and get a clean rip).
My issue, and I believe Scott's too is that people quote and requote the shure advertorial and no one other than ChannelD and Scott seem to have checked if the assertions are right. In a system like mine, with a Digital crossover and biquads to burn there are good reasons to stick an ADC in the phono stage and do the RIAA in the digital domain. It's not for everyone I grant but wouldn't the world be boring if we all implemented the same thing.
And I fully expect pitchforks and torches if I ever get the phantom powered MM headamp working
My issue, and I believe Scott's too is that people quote and requote the shure advertorial and no one other than ChannelD and Scott seem to have checked if the assertions are right. In a system like mine, with a Digital crossover and biquads to burn there are good reasons to stick an ADC in the phono stage and do the RIAA in the digital domain. It's not for everyone I grant but wouldn't the world be boring if we all implemented the same thing.
And I fully expect pitchforks and torches if I ever get the phantom powered MM headamp working
Clipping is not a problem, after all you don't need hifi to reproduce the click and pops. Recovery from clipping can be, but since Scott is using a no feedback approach (legitimate for a head amp) that's a non issue as well.
There is another confusion here too. We are talking about both digital and analog RIAA intermixed.
I followed the Pure Vinyl folks from the beginning and compared that to the simple out of hand comments that you can't do that it needs 40-50dB of headroom in the pre-amp. I published all the results, a USB A/D or field recorder with decent low noise mic pre-amps makes fine LP rips (MM/MI). I did need a small battery powered FET source follower as an impedance converter. The mini-DSP easily runs the RIAA as an IIR filter real time.
For analog I took Bob Cordell's idea and reduced it to one stage as a fun project. It is an open-loop V to I so the gain is determined by the impedance of the RIAA network, not the same as a fixed gain and passive RIAA. The FET's I used were high Vp so again the noise was appropriate for MM/MI and were still 0.01% linear at well over +-125mV at the input.
Walt and I built an instrumentation amplifier based pre-amp that was a fixed gain of 40dB at the input so that was around +-120mV and I never noticed any problems.
I guess there could be an issue with one size fits all, but I always try to set the gain for the cart in use. With the mic pre-amp on most of the recorders I have I suppose you could get to a gain where the SNR is not optimum but factoring surface noise in I have never had an issue (talking about the digital RIAA here).
There's nothing wrong with uber-designs with 0.3nV of noise and -120dB THD at full output, just offering something different for DIY entertainment.
Last edited:
Here is a link to the Holman articles Technical Requirements of Phono Preamplifiers by Tomlinson Holman
That's probably part of the disagreement. DIY, no problem tailoring for your very own cartridge. A product, not so much, at least without a very expensive solution (recall JC gain wire wound gain pot). Providing a large overload margin is definitely cheaper.
This IS a DIY forum after all. Doing things that could never be commerical products is where the fun is for me. A couple of transistors at the turnable phantom powered in a focusrite Scarlett is a potentially neat way of side stepping a lot of problems that analog front ends can have. A lot of people do rip vinyl. In fact some buy an LP and only play it once.
All good fun, and no bybees were hurt in the process
All good fun, and no bybees were hurt in the process
billshurv:
what a great "coke-spitter" that was.
i'm still wiping my keyboard
mlloyd1
what a great "coke-spitter" that was.
i'm still wiping my keyboard
mlloyd1
...
no bybees were hurt in the process
Of course, I was only trying to figure out where Tom Holman's opinion was originating from. TH didn't strike me as a DIYer
.
TH didn't strike me as a DIYer
No, but I wish he normalized all his data to one standard. Taking his last (worst case) graph I get 120mV for my 4mV cart, which is actually no problem except at the low end (maybe).
I wish there were more discussions of gain structure and overload margin, folks still complain when they use a miniDSP in front of their 106dB sensitivity speakers and find it noisy. They are running everything throwing away bits to get the output level low.
There is also the issue of overload at the input vs the output. I still don't see the point in specifying a line stage at 10V rms out when any normal 100 or so Watt PA would be hard railed at 3-4V.
Another thing about Mr. THX he talks about intermod with warp and LP eccentricity but this is not a pre-amp issue except in the extreme. The damage is done by the tracking angles and effective velocities changing, no amount of low pass filtering can fix that.
Last edited:
Yeah, there is though a good reason for that... traditionally, power amplifiers have clipping protections/flashing indicators/etc... Preamplifiers, not so much. It's probably a good idea to have a "single point of clipping", once again, from a consumer perspective. Of course, this is no good reson to feed a preamp at +/-24V and to make it almost rail-to-rail output (like, ahem, the Blowhard preamp).
10V RMS Line stage OL is specmanship IMO. 2V RMS out will cover just about anything on the planet. But I think if you offered a product that overloaded at 2 or 3V out, you would get flak for it. Some things are best just left where they are. That said, the distortion sweet spot on a lot of gear is the 2-3V output level - just look at the AP plots on Stereophile of the more sensible stuff.
With overload and the phono EQ, you have the volume control normally between the EQ OP and the power amp or line stage preamp so having good OL on the EQ amp allows you to essentially adjust the signal chain gain at a convenient place to fit the program material dynamics.
Even for DIY preamps, I'd still design the phono stage for good OL.
(separately, I've just been loaned a Hanna ML moving coil cart - absolutely superb and a big step up from the Ortofon Red I've been using)
With overload and the phono EQ, you have the volume control normally between the EQ OP and the power amp or line stage preamp so having good OL on the EQ amp allows you to essentially adjust the signal chain gain at a convenient place to fit the program material dynamics.
Even for DIY preamps, I'd still design the phono stage for good OL.
(separately, I've just been loaned a Hanna ML moving coil cart - absolutely superb and a big step up from the Ortofon Red I've been using)
The physical movement of two dissimilar insulators causing charge to move. Tribolectric effect.
Testing for it by the folks at Molex:
Triboelectric Noise in Medical Cables and Wires - The Connector: By Molex
Bohoho, waa, waa, sob, sob, nobody is in a rush to loan me such expensive stuff
. That must be one great cartridge.