I think there is an additional concern over catastrophic failure and huge DC currents as a result. Frankly since dual supply circuits have become common I can find few that cap couple phono inputs, I never even considered it. Heavy coat needed today, we are in complete snow white out right now.
Sure. But input stage overload @ 20Hz shouldn't be a problem for any decent RIAA circuit. It's preventing overloading @ 75us that's the main issue.
Yeah the bigger the better, no doubt. Unless we go for the "truly EVIL Ib cancelling schemes" 😀 as i have done with for eg OP27's etc
Have you simmed it ? I have & it works as one !
@ kgrlee
Why are you against Ib cancelling schemes for phono ? I havn't heard Any issues at all with mine.
Presumably you mean the circuit damaging the cartridge under fault conditions, though the converse is just about possible I suppose? But pretty unlikely in both cases, methinks. Different for commercial products versus DIY risks and consequences I think.
Just sayin' that it's not taboo by any means to pass a small, sane dc current through MM coils IME. Universe won't collapse any sooner than it will anyway, and fresh circuit ideas open for fun and sometimes advantage.
Keep warm and safe in the snow !
LD
A controlled small DC current slightly larger than cartridge normal condition maximum output current may not be a bad thing.
This class A current would avoid current reversal in the cartridge windings which may actually be advantageous.
Dan.
This class A current would avoid current reversal in the cartridge windings which may actually be advantageous.
Dan.
It's obvious you spurn 'real life' measurements & tests which don't confirm your own prejudices.
I've done similar tests like luckythedog on both MM & MC dating back to 1980 & before. As a real transducer guy, I can also spout loadsa theory on this.
Both the theory & practical tests say there is NO detectable difference ... but I'm certainly not going to try to persuade you. 😀
I'd rather use my time persuading you to do a worked out example of the JLH type 12dB/8ve filter around a single OPA RIAA stage. 🙂
If you are using OP27, here's something you can try. (I have.)Zero D said:
Measure the noise of your RIAA preamp with a real MM cartridge. Easy to do an FFT scan these days.
Replace your OP27 with cheapo NE5534. Check that feedback & output electrolytics are still properly biased and if compensation needs attention. Address these as required. Repeat your noise measurement & spectrum.
Some of the datasheets (including OP27 IIRC) explain what you will find. An RIAA preamp is about the best example of this EVIL behaviour.
Oh! You might find it useful to listen to the noise too. 🙂
Last edited:
Are you saying designers don't read datasheets, or can't understand what's in them? I could also be they don't know what's going to be at the input so they spec noise with a short, all of it fairly mediocre engineering.
Last edited:
I am using a bipolar electrolytic with a 0.47uF film in parallel to couple the cart to the input stage.
My concerns are that if the front end pops for some reason (not typical, but static can damage bipolar and JFET input devices as well) you run the risk of smoking your cartridge. If some dear enthusiast has a $500 cart (not unusual I might add), that's a pretty expensive problem.
You can pass a few mA through a cart and not hear the difference - I have not tried it, but believe you. However, you can also couple the cart via a 100 uF bipolar cap and not hear any difference. The AP also agrees with the latter BTW 😉
My concerns are that if the front end pops for some reason (not typical, but static can damage bipolar and JFET input devices as well) you run the risk of smoking your cartridge. If some dear enthusiast has a $500 cart (not unusual I might add), that's a pretty expensive problem.
You can pass a few mA through a cart and not hear the difference - I have not tried it, but believe you. However, you can also couple the cart via a 100 uF bipolar cap and not hear any difference. The AP also agrees with the latter BTW 😉
Yes, of course I simmed it.
A filter has a recognised response, such as Butterworth, Bessel, Chebyshev. They are used not to honour dead mathematicians but because the response is optimised in some way. Thus, the Butterworth has the fastest possible roll-off with a flat passband, the Cheb gives you faster roll-off if you are prepared to accept a tightly defined amplitude ripple in the passband, and so on.
Your circuit peaks at +0.35 dB before roll-off. The filter slope is degraded from 18 dB/oct to 12 dB/oct below -30 dB, in much the same way as the so-called JLH 'bootstrap filter' which was nothing of the sort. And a gain of +20dB means it is going to be very hard to fit into most signal paths.
Sorry if this seems negative, but you invited us to look at it. It's not a filter.
@ kgrlee
Hmmmm. I see what you're saying, but i've Never heard Any hiss/hum/noises etc from it, & i don't always listen @ low volume. Maybe if i did measure as you suggest it would show a difference, but i can't hear Anything bad, just the opposite. In fact it's Extremely revealing, & if i heard anything wrong i would have changed it. Maybe i'm deaf 😀
@ DouglasSelf
OK 😉
Well it doesn't have to be +20db.
I don't mind at all, i appreciate you looking into it 🙂 If we don't ask. we don't find out 😉 Back to the drawing board 😀
Hmmmm. I see what you're saying, but i've Never heard Any hiss/hum/noises etc from it, & i don't always listen @ low volume. Maybe if i did measure as you suggest it would show a difference, but i can't hear Anything bad, just the opposite. In fact it's Extremely revealing, & if i heard anything wrong i would have changed it. Maybe i'm deaf 😀
@ DouglasSelf
OK 😉
Well it doesn't have to be +20db.
I don't mind at all, i appreciate you looking into it 🙂 If we don't ask. we don't find out 😉 Back to the drawing board 😀
My OP27 example was a commercial amp. They had problems sourcing the OP27 so I said just swap it for a cheapo 5534.
Noise improved significantly. IIRC, nearly 3dB and the spectrum of the noise improved too.
There's no problem using something that's more $$$ and gives slightly worse performance. This isn't life support gear. Gotta cater for the deaf Golden Pinnae brigade if you are a commercial operation.
Better still is to use cheapo 5534, file off the designation and get some "hand made by Virgins from solid Unobtainium" labels to stick on top.
Then you get better performance AND Golden Pinnae cred too 🙂
But never confuse what you are doing to actually improve performance with what you are doing for da deaf Golden Pinnae. 😀
Different strokes, using an Ib comp'd in-amp as an RIAA input lifting the stylus off the groove always reduced the noise considerably this issue is a big don't care IMHO. In other words the surface noise of even the best LP's dominates. Using a cart as a source termination is not enough you have to be running in the best groove you can find.
Last edited:
I've a limited amount of data that suggests for certain types of MC & MM cartridges, the spectrum of the electrical noise .. always way below the surface noise as Guru Wurcer says ... may affect BLT results.
But this was 30+ yrs ago and I'm unlikely to ever do another vinyl playing system as my record collection is still in my sister's shed on the other side of the continent.
Which EVIL in-amps are these 😱
I remember looking at the specs of AD797 when it came out and thinking how clever these AD guys must be to have big current in the input stage and small Ib.
I was shocked when Guru Wurcer told me his iconic AD797 had EVIL Ib cancelling 😱
Duu.uuh! As the only sensible use of an in-amp is with a floating or balanced source, Ib compensation isn't really EVIL for them.
... unlike their truly EVIL use in unbalanced amps with very different Z on the +ve & -ve inputs. 🙂
This is the crux, the elephant in the room, and is true, of course.
I've never understood the quest for noise performance of op-amps in vinyl preamp circuits. The humble 5534 surely ticks the boxes for all but the most exotic of circuits IME. This, plus Ib being a tolerable issue for most circuits, though Ib cancelling prob being net bad, makes it easy.
LD
@ scott wurcer & kgrlee
About 15 years ago i designed & built a small mixer/preamp for my home studio. One input was for a beyer mic. I used the AD797, with a large uf capacitor on the input, & am extremely pleased with it 🙂 Again, i hear no noise etc
About 15 years ago i designed & built a small mixer/preamp for my home studio. One input was for a beyer mic. I used the AD797, with a large uf capacitor on the input, & am extremely pleased with it 🙂 Again, i hear no noise etc
Who said otherwise? It's when someone simply HAS to have a beeswax and silver-foil $400 input cap, in that case leave it out.
Err.rh! What I said is "I've a limited amount of data that suggests for certain types of MC & MM cartridges, the spectrum of the electrical noise .. always way below the surface noise as Guru Wurcer says ... may affect BLT results."
I was a DBLT guru in my previous life. I divide all performance aspects into stuff that's either audible in DBLTs or NOT.
I'm as prone to chasing 1ppm THD20k as any on this forum but I'm am well aware that this is just wanking. 🙂
The tests I mentioned weren't Double Blind (really difficult on a vinyl system) and there were other things that could explain the preference for the lower noise presentations.
It's finding out the reasons for these preferences which may lead to an increase in the fund of human knowledge.
Then there's stuff which is bleedi'n obvious in sighted tests which disappear completely on a DBLT ... which is why JC & other luminaries refuse to do ANY DBLTs. 😀
5534 is actually nearly ideal for MM cartridges. You won't get even 1dB better noise with anything more $$$ .. that's if you get better noise at all.
Although I don't think you are trying to call Douglas Self a liar, you have contradicted a statement in his book:
In practical measurements with a 5534A as amplifier A1, I found that the noise improvement [using external circuit X] with a real cartridge load (Shure M75ED 2, cartridge parameters 610 ohms + 470 mH) was indeed 1.3dB, just as predicted
I wasn't complaining etc, just the opposite 😉
Unless you can get a good deal on dem caps, eg $360 😀