Quite possibly. I also built it at the time. It nearly put me off diy for life. Ok, i had no experience with opamp 'sound' at the time and the phono section of my NAD3150 played so much better...
Later i discovered that the 5534 is only acceptable in inverting mode - not ideal for phono - and the split passive/active correction will never sound good to me.
Best phono preamp structure
In ETI 468 MM is used particularly (infrequently) phono preamp structure.
You could find several reasons justifying actual used structures in
http://www.national.com/an/AN/AN-346.pdf
As well as detailed calculus about how to compute component's values involved in one of the most actually RIAA preamplifier structure.
Implementing 75 microseconds pole early in the first stage is not best solution concerning noise …
In ETI 468 MM is used particularly (infrequently) phono preamp structure.
You could find several reasons justifying actual used structures in
http://www.national.com/an/AN/AN-346.pdf
As well as detailed calculus about how to compute component's values involved in one of the most actually RIAA preamplifier structure.
Implementing 75 microseconds pole early in the first stage is not best solution concerning noise …
True. Is noise really such an important issue? Some of the phono pres i've built have been dead-quiet (LM394, INA103 inputs) but possibly not so good sounding. The best opamp topology IME is a 3 stage circuit with split passive correction. A Hagerman Buggle is an example of the basic topology.
analog_sa said:
Could you tell me where I can see it?
I'd also like to know if it is a good idea to make an input buffer, I mean put a low current noise op amp at the input and then use a passive RIAA ecualization and then add another op amp:
input buffer - passive RIAA rc network - gain (maybe with hi pass filter for rumble).
Thanks.
I think it would be better to give some gain at the input instead of at the output. The way you would do it you would amplify the noise picked up in the previous strages. Giving some (or all) the gain in the first low noise stage has the advantage that the RIAA decreases the amplitude at higher frequencies. This way you attenuate the HF noise.
rho said:
Yeah! I was thinking about that too. Making the first op amp (at the input) to deliver high gain, then the passive RIAA and then the output opamp with slight gain... would it be a good idea to implement the RIAA equalization (for less deviation from the curve) in two parts, one with passive network and the other at the output opamp (without the rumble filter)?
Which opamp should I use at the input??? OPA627 (low current noise)? Or maybe there are some other good opamps for that? Thanks again.
best RIAA phono on ICs
Hi All,
Firstly, Analog, it is no wonder that ETI 478MM in its standard form disappointed you but DIY approach, among other things, allows you for some experimentation. Inverting phono inputs unfortunately are noisy and therefore should be avoided in MM/MC first stages.
The first mods one could do with ETI 478 are: get rid of input and output electrolytes; use a BP paralelled by a polypropylene; use high quality resistors and metallized polypropylene caps in the circuit and pay attention to good filtering - at least 220nF across +/-V pins of each IC and at least hundreds, if not thousands, of uF bypassed by 100-220nF/100V polypropylene/polycarbonate (were available in those days) caps for each IC on their pins down to the ‘dirty’ ground plane. If you had only MM pickup, as I had, the phono input should be connected directly to the input sockets and not to the switch. The ETI-5000 had far too much of screened cabling and switches.
The second modification could involve IC upgrades as OP37 became available in mid eighties. Later, of course, much better IC appeared on the marked of which LT1028 could be a good choice for the first IC if one wanted noise to be minimized and OPA627BP (or OPA637BP) for the second IC. The ‘noise level’ equations included in my comments are worth consideration. Note that Boltzman’s constant is 1.38065 etc but the typo is academic in this case. Also it is ETI 478 not 468. Looks my key punch precision is sometimes lacking.
In my opinion there are 3 main areas of concern when designing a RIAA preamp, namely:
noise (proper choice of OP-AMPs or transistors or a ‘mixture’ of both);
interaction with the cartridge should be minimised while RIAA circuit should not only use high quality components but should be fine tuned as each of its stages interacts with all the other; and that may be more audible than deviations from RIAA up to even 0.5dB except for the 300Hz – 6kHz range where our hearing is most sensitive and errors should be at least within +/-0.2dB (don’t forget the dilemma: larger resistor or a capacitor – noise-delays-phase shifts); and
thirdly, power supply and on board filtering should be low noise and adequate.
The rest results from the above and forget about one stage RIAA preamp if one is after quality.
Now, Bucurb, the 75us time constant, which defines a low pass filter with –3dB at 2122Hz as you know, should be used rather early in the pre-amp as it allows high frequency overload capability to be shaped properly before the next stage(s). Of course it is a compromise but probably a better one than most other solutions. I’ll add here that implementing it as a passive stage has the advantage of ensuring that the impedance seen by the cartridge is constant with frequency, which might become crucial in some designs.
Each design concept has its advantages and disadvantages. Choosing a passive RIAA solution has a few advantages as well a couple of disadvantages of which the most prominent is higher noise level than in a well designed two stage mixed (passive/active) equalisation preamp. The other problem with the passive approach is that the RIAA circuit requires a lot of fine tunning as stages interact with each other. That is why some choose a three stage design as this one allows for better separation of RIAA stages and reduces their interactions.
To the contrary of what seems a popular belief among some people designing a good passive RIAA circuit is quite demanding. For all considerations see Stanley P. Lipshitz – “On RIAA equalisation networks” in Journal of the Audio Engineering Society, June 1979 or read the article “Designing valve preamps” by Morgan Jones in Electronics World. March 1996.
When it comes to a passive RIAA solution, judging by data sheets, I think the best ICs for the first stage include: THAT1512 and INA217 followed by THAT1510 and SSM2019. The following stages are less demanding but OPA627BP/637BP are probably among the best. An alternative is a discrete design or a mixed one with the latter being probably a good compromise. Here I must confess that I am not a vinyl fanatic but still have heaps of vinyl not available on CDs.
I must add that I am tempted to build a preamp with passive RIAA and find out which of the two designs I like most in direct comparison. Now guys, has any of you had a chance to directly compare the best passive RIAA with best mixed two stage RIAA preamps? What are the subjective advantages and drawbacks of each of these designs?
Cheers,
Hi All,
Firstly, Analog, it is no wonder that ETI 478MM in its standard form disappointed you but DIY approach, among other things, allows you for some experimentation. Inverting phono inputs unfortunately are noisy and therefore should be avoided in MM/MC first stages.
The first mods one could do with ETI 478 are: get rid of input and output electrolytes; use a BP paralelled by a polypropylene; use high quality resistors and metallized polypropylene caps in the circuit and pay attention to good filtering - at least 220nF across +/-V pins of each IC and at least hundreds, if not thousands, of uF bypassed by 100-220nF/100V polypropylene/polycarbonate (were available in those days) caps for each IC on their pins down to the ‘dirty’ ground plane. If you had only MM pickup, as I had, the phono input should be connected directly to the input sockets and not to the switch. The ETI-5000 had far too much of screened cabling and switches.
The second modification could involve IC upgrades as OP37 became available in mid eighties. Later, of course, much better IC appeared on the marked of which LT1028 could be a good choice for the first IC if one wanted noise to be minimized and OPA627BP (or OPA637BP) for the second IC. The ‘noise level’ equations included in my comments are worth consideration. Note that Boltzman’s constant is 1.38065 etc but the typo is academic in this case. Also it is ETI 478 not 468. Looks my key punch precision is sometimes lacking.
In my opinion there are 3 main areas of concern when designing a RIAA preamp, namely:
noise (proper choice of OP-AMPs or transistors or a ‘mixture’ of both);
interaction with the cartridge should be minimised while RIAA circuit should not only use high quality components but should be fine tuned as each of its stages interacts with all the other; and that may be more audible than deviations from RIAA up to even 0.5dB except for the 300Hz – 6kHz range where our hearing is most sensitive and errors should be at least within +/-0.2dB (don’t forget the dilemma: larger resistor or a capacitor – noise-delays-phase shifts); and
thirdly, power supply and on board filtering should be low noise and adequate.
The rest results from the above and forget about one stage RIAA preamp if one is after quality.
Now, Bucurb, the 75us time constant, which defines a low pass filter with –3dB at 2122Hz as you know, should be used rather early in the pre-amp as it allows high frequency overload capability to be shaped properly before the next stage(s). Of course it is a compromise but probably a better one than most other solutions. I’ll add here that implementing it as a passive stage has the advantage of ensuring that the impedance seen by the cartridge is constant with frequency, which might become crucial in some designs.
Each design concept has its advantages and disadvantages. Choosing a passive RIAA solution has a few advantages as well a couple of disadvantages of which the most prominent is higher noise level than in a well designed two stage mixed (passive/active) equalisation preamp. The other problem with the passive approach is that the RIAA circuit requires a lot of fine tunning as stages interact with each other. That is why some choose a three stage design as this one allows for better separation of RIAA stages and reduces their interactions.
To the contrary of what seems a popular belief among some people designing a good passive RIAA circuit is quite demanding. For all considerations see Stanley P. Lipshitz – “On RIAA equalisation networks” in Journal of the Audio Engineering Society, June 1979 or read the article “Designing valve preamps” by Morgan Jones in Electronics World. March 1996.
When it comes to a passive RIAA solution, judging by data sheets, I think the best ICs for the first stage include: THAT1512 and INA217 followed by THAT1510 and SSM2019. The following stages are less demanding but OPA627BP/637BP are probably among the best. An alternative is a discrete design or a mixed one with the latter being probably a good compromise. Here I must confess that I am not a vinyl fanatic but still have heaps of vinyl not available on CDs.
I must add that I am tempted to build a preamp with passive RIAA and find out which of the two designs I like most in direct comparison. Now guys, has any of you had a chance to directly compare the best passive RIAA with best mixed two stage RIAA preamps? What are the subjective advantages and drawbacks of each of these designs?
Cheers,
Could somebody tell me where can I order hi quality parts in the US? Any online store out there? I live in a very small town near United States so here I can't buy nothing but standard caps, resistors, and ICs... I'd like to know where to get 1% resistors and hi quality caps on the net! Thanks.
quality caps and resistors,
Hi Alexei,
Two good sites for these goodies in the US are: Micael Percy Audio and Handmade Electronics.
Michael's site address: www.percyaudio.com
Download their catalogue from the site and then email Michael on: mpercy@pacbell.net to discuss and place your order.
Handmade Electronics site is here: www.hndme.com
Most of the other parts you can get from Mouser and DigiKey.
Cheers,
Hi Alexei,
Two good sites for these goodies in the US are: Micael Percy Audio and Handmade Electronics.
Michael's site address: www.percyaudio.com
Download their catalogue from the site and then email Michael on: mpercy@pacbell.net to discuss and place your order.
Handmade Electronics site is here: www.hndme.com
Most of the other parts you can get from Mouser and DigiKey.
Cheers,
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.