These all look like good alternatives:
Texas Instruments | Voltage Reference, 2.5V |LT1009CLPR
Texas Instruments | Voltage Reference, 2.49V |LM336LP-2-5
Linear Technology | Voltage Reference, 2.5V |LT1009CZ#PBF
Similar specs, nice low noise etc. Just leave the "adjust" pin unconnected.
Don't see anything useful at Maplin, though.
Looking at that Head-Fi thread now...
Hi Godfrey,
Nice find😀 My only concern is that the ZN458B is described as "2.45V PRECISION REFERENCE REGULATOR". Presumably it does't matter that the ones you've found are higher voltage than that? 😕
I see that RS also do polystyrene caps (much cheaper than I had previously found). The ones for this project are listed as "1% polystyrene caps", but I've been unsuccessful in finding out what the 1% means. Is it relevant ?
One last capacitor question for you. I saw 50V mentioned in the article, so I've been working on the assumption that all the caps I need must be rated at least 50V unless otherwise stated. Is this anywhere near the truth ?
Sorry for the number of questions, but I should probably get this straight before I go ordering anything else !
Thanks as always,
Tim
Nice find😀 My only concern is that the ZN458B is described as "2.45V PRECISION REFERENCE REGULATOR". Presumably it does't matter that the ones you've found are higher voltage than that? 😕
I see that RS also do polystyrene caps (much cheaper than I had previously found). The ones for this project are listed as "1% polystyrene caps", but I've been unsuccessful in finding out what the 1% means. Is it relevant ?
One last capacitor question for you. I saw 50V mentioned in the article, so I've been working on the assumption that all the caps I need must be rated at least 50V unless otherwise stated. Is this anywhere near the truth ?
Sorry for the number of questions, but I should probably get this straight before I go ordering anything else !
Thanks as always,
Tim
Hi Tim
Re the voltage references - I think the difference is trivial. Since Andy's roughly doubling the original voltage and the parts he specifies are only accurate to +-1.5% anyway, I don't think +2% will make any difference.
For capacitors, the "1%" is the tolerance.
e.g. a 1nF 1% capacitor could be anywhere between 0.99nF and 1.01nF whereas a 1nF 5% capacitor could be anywhere between 0.95nF and 1.05nF.
C115 and C116 (the 1000pF ones) are quite critical. They (together with the 75K resistors) determine the high-frequency roll-off for the RIAA equalization. So e.g. if those caps are 10% too high, all of the treble (above about 2kHz) will be reduced about 1dB, but if the caps are 5% low, you'll get about 1/2 a dB too much treble.
C103 and C104 (the 680pF ones) aren't critical at all - they're just there for stability. If in doubt or struggling to find exact values, rather use a little higher cap value than a little lower.
Re capacitor voltage ratings:
I presume the opamps are running off +-15V supplies so a rating of anything over about 30V should be fine for the polystyrene caps.
The 2.2uF polypropylene caps shouldn't have much (if any) voltage across them so probably any rating will do. If you find higher value caps at a sensible price, that wouldn't hurt. OTOH, smaller isn't a disaster either. They help define the low-frequency roll-off (bigger is better). Note that Andy deliberately reduced the value (but increased the quality) to sacrifice a bit of low-end bass extension in exchange for (hopefully) better sound quality.
The 100uF electrolytics are another kettle of fish. Frankly, I don't know why Andy bothered since the impedance of the voltage regulators is likely below the impedance of the caps throughout the audio band so the caps won't do anything useful anyway.
However... let's go with the flow:
They'll have about 2.5 volts across them (if used) so the voltage rating doesn't really matter. However, if they're to do anything useful, they need to be low-ESR (equivalent series resistance) types. The exact value actually isn't important - High capacitance is good and low ESR is good. If you're comparing datasheets for caps, look for ones with low-esr or low impedance (at whatever frequency is specified). A few rules of thumb:
a) caps with higher voltage rating often have lower ESR.
b) caps recommended for use in SMPS (switch-mode power supplies) are probably better.
c) caps with higher temperature ratings (e.g. 105C vs 85C) are probably better.
Cheers - Godfrey
Re the voltage references - I think the difference is trivial. Since Andy's roughly doubling the original voltage and the parts he specifies are only accurate to +-1.5% anyway, I don't think +2% will make any difference.
For capacitors, the "1%" is the tolerance.
e.g. a 1nF 1% capacitor could be anywhere between 0.99nF and 1.01nF whereas a 1nF 5% capacitor could be anywhere between 0.95nF and 1.05nF.
C115 and C116 (the 1000pF ones) are quite critical. They (together with the 75K resistors) determine the high-frequency roll-off for the RIAA equalization. So e.g. if those caps are 10% too high, all of the treble (above about 2kHz) will be reduced about 1dB, but if the caps are 5% low, you'll get about 1/2 a dB too much treble.
C103 and C104 (the 680pF ones) aren't critical at all - they're just there for stability. If in doubt or struggling to find exact values, rather use a little higher cap value than a little lower.
Re capacitor voltage ratings:
I presume the opamps are running off +-15V supplies so a rating of anything over about 30V should be fine for the polystyrene caps.
The 2.2uF polypropylene caps shouldn't have much (if any) voltage across them so probably any rating will do. If you find higher value caps at a sensible price, that wouldn't hurt. OTOH, smaller isn't a disaster either. They help define the low-frequency roll-off (bigger is better). Note that Andy deliberately reduced the value (but increased the quality) to sacrifice a bit of low-end bass extension in exchange for (hopefully) better sound quality.
The 100uF electrolytics are another kettle of fish. Frankly, I don't know why Andy bothered since the impedance of the voltage regulators is likely below the impedance of the caps throughout the audio band so the caps won't do anything useful anyway.
However... let's go with the flow:
They'll have about 2.5 volts across them (if used) so the voltage rating doesn't really matter. However, if they're to do anything useful, they need to be low-ESR (equivalent series resistance) types. The exact value actually isn't important - High capacitance is good and low ESR is good. If you're comparing datasheets for caps, look for ones with low-esr or low impedance (at whatever frequency is specified). A few rules of thumb:
a) caps with higher voltage rating often have lower ESR.
b) caps recommended for use in SMPS (switch-mode power supplies) are probably better.
c) caps with higher temperature ratings (e.g. 105C vs 85C) are probably better.
Cheers - Godfrey
Ah... I see, that makes sense. Should save me a lot of time and money - cheers for that. I'll start paying closer attention to component tolerances now you have pointed this out.
Thank you, I had a suspicion this may have been the tolerance, but I couldn't find anything to confirm it. Even so, I would not have understood the significance - so thanks again.
Is there any "assumed" tolerance for caps that do not specifically state a tolerance, or should I avoid any like that?
Something else I didn't know!
Excellent - thank you.
This is very surprising for me, but answers a lot of questions about other projects I've looked at. So are you saying that it literally doesn't matter how high I go with the value, as long as it isn't at the expense of quality?
Not sure I understand this part, but I will take a proper look at it in the context of the circuit diagram when I get home to see if I can make more sense of it. You may have just found my limit for the moment. Note to self: look up the meaning of "impedance" 😀
Yet more useful information. I think I'll be throwing away my current shopping list and starting again as a result of this little lot!
What interesting read, thank you very much indeed for taking the time to explain all that, Godfrey. I hope you don't feel I'm taking advantage of your helpful nature 🙂
Btw, I've found a reasonably priced supply of the AD743, but in what looks like a 16pin surface-mount package rather than 8 pin - which I presume are identical other than the pinouts.
Digi-Key - AD743JRZ-16-ND (Manufacturer - AD743JRZ-16)
If the opa627s I ordered turn out to be duff, I may well try these and have a go at adapting them to go into an 8 pin socket
Reasonable assumption: either it's embarrassingly bad or they just couldn't be bothered to print it or they have a box of mixed-up stuff and what you get is lucky dip.
Most parts actually do have a tolerance though, so if you've already got some, it's probably possible to figure out what they are. Sometimes it's explicit like "5%". Often a letter code e.g. "K" or "J" is used to indicate the tolerance, but I can't remember which is what with the letters. There may be other even more obscure markings as well. If you know the manufacturer, there's likely to be info on there website on how to decipher their codes.
For coupling caps, yes, within the bounds of reasonableness.
e.g. I see Maplin has some "Audio-Grade" polypropylene caps here:
Audio-Grade Polypropylene Axial Capacitors
Note that the largest value there (10uF) is nearly 2 inches diameter and 2 inches long (and costs 9GBP each). 😱 I'd think twice about that, let alone anything bigger.
Also; any clicks or pops e.g. at switch-on or when switching between sources on the pre-amp may be worse with bigger coupling caps.
Nice find!
Pity you didn't spot that a couple of days ago.
Regarding the 100uF capacitors:
They're supposed to be reducing what little noise there is from the voltage reference. Using a capacitor and a resistor (as in the third pic below) would make a nice filter and reduce the noise substantially.
Just the cap on it's own isn't going to have much effect because then the filter is made of only the cap and the "built-in" resistance of the voltage reference, which is something less than one ohm.
For filters like that, bigger is better for the cap and the resistor. e.g. Increasing that 100R resistor to 1K would likely be a big improvement. You can't go too high though or it'll start affecting the regulation.
One problem with adding parts is it gets messy when they don't fit on the PCB.
I don't suppose Andy's a member on this forum? It would be nice to ask his opinion and find out if there's good logic there or if he was just having a "blond" moment. (happens to all of us 😀)
Cheers - Godfrey
Attachments
Indeed 🙁
Do you think there'd be much benefit in replacing the NE5534 with AD743 (as opposed to recycling the AD744) ? At this price, I'm tempted.
I might try your improved filter too, if I ever get this finihsed !
Last edited:
Dunno. I haven't used any of these (OPA627/AD743/AD744), so I can't comment on the sonics.
If you do get the 743s as well as the 627s, you can always try swapping things around to see what works best. I suppose first thing is to see how the 627s turn out.
What would be really interesting is if you do the mods one at a time, to get an idea of how much difference each one makes. i.e. input stage mods vs chip-swap vs coupling-cap swap.
From a technical pov, I'd expect the input stage mods to make the most difference and the output coupling caps to have the least effect, but I'd be interested to hear about the subjective effects on sound quality.
btw, regarding the output caps, IIRC the idea is to swap black gate electrolytics for polypropylenes, but maybe you can get away with leaving them out altogether (following the "best cap is no cap" philosophy) - just replace with a little wire link e.g. a cut-off from a resistor leg or something. Worst that can happen is a loud pop when switching between sources on the pre-amp. probably best to try it at low volume first just in case.
Cheers - Godfrey
Hi Godfrey, I've ordered all the parts, but i'll hold off getting the 743s for now. I'll follow your suggestions and let you know how it goes.
Thanks again for all your help and advice. You never know, this might be the beginning of a new hobby for me 🙂
Bye for now
Tim
Thanks again for all your help and advice. You never know, this might be the beginning of a new hobby for me 🙂
Bye for now
Tim
Update
Hi Godfrey, just though I'd let you know how I'm getting on.
Over the weekend, I replaced C121/C122 with the two Solen 2.2uF Polypropelenes, and the two resistors R121/R122 to increase the Gain.
( Sorry to say, I didn't have the nerve to try your suggestion of completely shorting out the terminals for C121/122 - hope you're not too disappointed in me 😱 )
I did play a couple of tracks after swapping the caps to make sure the thing was still working, but I then rushed straight into chainging the two resistors - so I didn't really do a proper evaluation of the cap upgrade on its own.
What I can tell you though is that something has made a huge change for the better (I guess the caps). Everything sounds silky, with (seemingly) more fine detail. I'm now playing and enjoying albums that I previously had thought were less than top notch recordings. Turns out it was my phono amp making them sound grainy!
Wish I'd done this years ago. The only down side is the too many late nights sitting up enjoying my vinyl collection 🙂 Hard to believe that this part of the mod has only set me back < 8GBP! Andy certainly knows his stuff.
The two OPA627 chips have just arrived today from Hong Kong (at least, I hope that's what they are 🙁 ). I'll put these in at some point soon at see what happens.
Still waiting for some polystyrene capacitors to turn up before I can do the rest of the work.
- Tim
Hi Godfrey, just though I'd let you know how I'm getting on.
Over the weekend, I replaced C121/C122 with the two Solen 2.2uF Polypropelenes, and the two resistors R121/R122 to increase the Gain.
( Sorry to say, I didn't have the nerve to try your suggestion of completely shorting out the terminals for C121/122 - hope you're not too disappointed in me 😱 )
I did play a couple of tracks after swapping the caps to make sure the thing was still working, but I then rushed straight into chainging the two resistors - so I didn't really do a proper evaluation of the cap upgrade on its own.
What I can tell you though is that something has made a huge change for the better (I guess the caps). Everything sounds silky, with (seemingly) more fine detail. I'm now playing and enjoying albums that I previously had thought were less than top notch recordings. Turns out it was my phono amp making them sound grainy!
Wish I'd done this years ago. The only down side is the too many late nights sitting up enjoying my vinyl collection 🙂 Hard to believe that this part of the mod has only set me back < 8GBP! Andy certainly knows his stuff.
The two OPA627 chips have just arrived today from Hong Kong (at least, I hope that's what they are 🙁 ). I'll put these in at some point soon at see what happens.
Still waiting for some polystyrene capacitors to turn up before I can do the rest of the work.
- Tim
Last edited:
I've just had my meter out on the chips that arived today, and I'm getting a reading that seems to imply - from the above thread, at least - that they could well be genuine. 50k on one chip, 58k on the other. 🙂
Some of our Members put down every Journalistic and every Reviewer as if they can only spew trash.
Andy Grove was part of HiFiWorld from it's early days and contributed a lot to that mag's reputation.
It's one of the few I relied on, but there are many competent reviewers out there. We just have to learn to recognise those who know what they are discussing.
Rely on Andy Grove. He is one of the good guys. It would be nice if he joined us here. Is he still with those valve/tube folk?
Andy Grove was part of HiFiWorld from it's early days and contributed a lot to that mag's reputation.
It's one of the few I relied on, but there are many competent reviewers out there. We just have to learn to recognise those who know what they are discussing.
Rely on Andy Grove. He is one of the good guys. It would be nice if he joined us here. Is he still with those valve/tube folk?
whoops
ok, so I replaced the AD744's with OPA627's in the input stage and moved the AD744s to the output stage, and everying was working fine and sounding very good indeed.
I then went ahead and carried out the remaining part of the mod which was to "reduce the noise at the input stage and allow the opamp more optimally by reducing the DC common mode input voltage"
Since doing this, with the 627s inserted in the input stage, I either get a clicking noise when adjusting the volume on the pre-amp, or I get a whining noise coming from a nearby speaker when I switch the phono amp on (in the case of the latter, the phono-amp seems to be transmitting this noise wirelessly to the speaker 😕). If I replace the 627s wth something like the OPA1641, or the original AD744s, things seem to work ok. In fact, the 1641s sound pretty good.
Experimenting further, I found that by moving the OPA627's to the output stage, they seem to work fine, which makes me suspect that either something has gone wrong with this latest input stage mod, or there is something these chips dont like about it.
Can anyone help me diagnose this before I go ahead and reverse this latest mod?
I have ordered a capacitance tester in case I have blown any of the poly caps by overheating when soldering them - although I have no idea whether a blown cap would cause this problem.
Maybe my 627's are fake after all ?
Any help very much appreciated - as always.
Thanks - Tim
ok, so I replaced the AD744's with OPA627's in the input stage and moved the AD744s to the output stage, and everying was working fine and sounding very good indeed.
I then went ahead and carried out the remaining part of the mod which was to "reduce the noise at the input stage and allow the opamp more optimally by reducing the DC common mode input voltage"
Since doing this, with the 627s inserted in the input stage, I either get a clicking noise when adjusting the volume on the pre-amp, or I get a whining noise coming from a nearby speaker when I switch the phono amp on (in the case of the latter, the phono-amp seems to be transmitting this noise wirelessly to the speaker 😕). If I replace the 627s wth something like the OPA1641, or the original AD744s, things seem to work ok. In fact, the 1641s sound pretty good.
Experimenting further, I found that by moving the OPA627's to the output stage, they seem to work fine, which makes me suspect that either something has gone wrong with this latest input stage mod, or there is something these chips dont like about it.
Can anyone help me diagnose this before I go ahead and reverse this latest mod?
I have ordered a capacitance tester in case I have blown any of the poly caps by overheating when soldering them - although I have no idea whether a blown cap would cause this problem.
Maybe my 627's are fake after all ?
Any help very much appreciated - as always.
Thanks - Tim
Tim,
I bought this RIAA a month ago and did the mods. After doing the last mod in the article, mine didn't like OPA627 in stage 1 either, it was whinning.
I tried LT1115, LME49710 and OPA627 in different combinations.
So far I settled with LME49710 for both stages.
LT1115 in stage 1 and LME49710 in stage 2 sounds good too.
Next I'll replace all resistors with Dale CMF55.
This RIAA sounds pretty good.
Preamp is Audio Research LS-3 and amp is LM4780 (diy). Denon DL-160 cartridge, using the MM input.
Cheers
Jan
I bought this RIAA a month ago and did the mods. After doing the last mod in the article, mine didn't like OPA627 in stage 1 either, it was whinning.
I tried LT1115, LME49710 and OPA627 in different combinations.
So far I settled with LME49710 for both stages.
LT1115 in stage 1 and LME49710 in stage 2 sounds good too.
Next I'll replace all resistors with Dale CMF55.
This RIAA sounds pretty good.
Preamp is Audio Research LS-3 and amp is LM4780 (diy). Denon DL-160 cartridge, using the MM input.
Cheers
Jan
Final choice of opamps was LME49710 in stage 1 and OPA627 in stage 2.
Using OPA627 in stage 2 makes it noise free, whereas any of the other opamps in stage 2 would make it hiss at maximum volume.
Jan
Using OPA627 in stage 2 makes it noise free, whereas any of the other opamps in stage 2 would make it hiss at maximum volume.
Jan
Hi Jan,
Very interesting to see your reply - thank you very much. I believe I did also end up with the OPA627 in stage 2, but I'm trying to remember what I had in stage 1! I went through 3 or four different options if I remember. Very happy with the sound now though.
I will take a look at some point and let you know for sure... it's a bit of turmoil here at the moment, so difficult for me to check 🙂
You've now got me wanting to try the LME49710 for comparison. Are they expensive?
Edit:
I've just noticed that you've soldered the ICs directly to the PCB. Did this have any effect? I still have mine in sockets.
Very interesting to see your reply - thank you very much. I believe I did also end up with the OPA627 in stage 2, but I'm trying to remember what I had in stage 1! I went through 3 or four different options if I remember. Very happy with the sound now though.
I will take a look at some point and let you know for sure... it's a bit of turmoil here at the moment, so difficult for me to check 🙂
You've now got me wanting to try the LME49710 for comparison. Are they expensive?
Edit:
I've just noticed that you've soldered the ICs directly to the PCB. Did this have any effect? I still have mine in sockets.
Hi Tim,
LM49710 are not that expensive. You can find them on Ebay or here.
After having done the last mod in the article (Long Tailed Pair) OPA627 didn't work for me in Stage 1. It was howling.
Last night after having found the opamp combination I liked the most, I soldered them directly to the pcb. Hard to tell if it had any affect, 'cause at the same time I replaced all resistors with Dale, and I installed Elna Starget 100uf/16V over the legs of R113/R115 & R114/R116 (first I had Nichicon Muse there).
Overall I'm quite happy with the sound, and I'm in no hurry for a replacement of this RIAA.
LM49710 are not that expensive. You can find them on Ebay or here.
After having done the last mod in the article (Long Tailed Pair) OPA627 didn't work for me in Stage 1. It was howling.
Last night after having found the opamp combination I liked the most, I soldered them directly to the pcb. Hard to tell if it had any affect, 'cause at the same time I replaced all resistors with Dale, and I installed Elna Starget 100uf/16V over the legs of R113/R115 & R114/R116 (first I had Nichicon Muse there).
Overall I'm quite happy with the sound, and I'm in no hurry for a replacement of this RIAA.
Next to the stage 2 opamps there are four empty places for 0.1uf decoupling caps (C123-C126).
Perhaps I will try some good 0.1uf polypropylenes there, to see if improves the sound. It's just a shot from the hip, and I don't have sufficient electronic skills to tell if and when decoupling are needed at this place...
Perhaps I will try some good 0.1uf polypropylenes there, to see if improves the sound. It's just a shot from the hip, and I don't have sufficient electronic skills to tell if and when decoupling are needed at this place...
Have now bypassed the OPA627 supply (Black Gate 100uf) with four 0.15uf ERO MKT 1822.
Didn't make things worse, for sure!
Didn't make things worse, for sure!
Last edited:
Reversed above, since they didn't add to anything.
Changed the 1000uf noname PSU caps to Nichicon KZ.
Changed the 1000uf noname PSU caps to Nichicon KZ.
- Status
- Not open for further replies.