I do not know where this person is getting their information from.
Which is a bunch of bunk, I have in front of me the original Signetics NE5532/5532A data sheet from a 1979 data manual. It is the original spec.
It says in the Features:
Large supply voltage range +/-3 to +/-20V
It is characterized at +/-15V
A +/-17V is not "probably alright", it is just fine, it is all in the wording.
No, this is the first time I have heard of this person. He is not the designer of this 2102 pre-amp. This is why I'd like to see his measurement results.
B.Duncan gets discussed here 12 years ago.
http://www.diyaudio.com/forums/analog-line-level/69974-ben-duncans-amp-01-a.html#post793150
http://www.diyaudio.com/forums/analog-line-level/69974-ben-duncans-amp-01-a.html#post793150
In reference to the Elektor project issued in April, May and June 1012: -
As i clearly stated the text very clearly says that the information was given to me direct from Elektor. As i said if you are not happy with that information then write direct to Elektor.
If you have the BOM list for the Elektor Preamplifier the i.c.'s are as NE5532ANG and NE5534ANG.
Elektor said to use the NE5532AP and NE5534AP, these being the 2017 types available.
In Farnell it very clearly says "Supply Voltage Range: ± 5V to ± 15V.
If you are not happy with that then write to Farnell!
Clearly we should be looking at modern day spec of the i.c.'s that are supplied now and not looking at spec of those i.c.'s issued 38 years ago back in 1979.
I concentrate on the write up and technical description in Elektor's 3-part Preamplifier 2012, issued in April, May and June 2012 and not from other sources.
For the Measurement results read these three parts and not other sources.
As i clearly stated the text very clearly says that the information was given to me direct from Elektor. As i said if you are not happy with that information then write direct to Elektor.
If you have the BOM list for the Elektor Preamplifier the i.c.'s are as NE5532ANG and NE5534ANG.
Elektor said to use the NE5532AP and NE5534AP, these being the 2017 types available.
In Farnell it very clearly says "Supply Voltage Range: ± 5V to ± 15V.
If you are not happy with that then write to Farnell!
Clearly we should be looking at modern day spec of the i.c.'s that are supplied now and not looking at spec of those i.c.'s issued 38 years ago back in 1979.
I concentrate on the write up and technical description in Elektor's 3-part Preamplifier 2012, issued in April, May and June 2012 and not from other sources.
For the Measurement results read these three parts and not other sources.
Last edited:
NE5532ANG and NE5534ANG are the part numbers from Onsemi. They are still in production.
NE5532AP and NE5534AP are the TI part numbers, also still in production.
The NE5532ANG and NE5534ANG are recommended by Doug Self, thus the ones called out in the BOM because he measured the two sources and found slightly better performance of the Onsemi versions. I'd have to look to determine exactly which article he did this comparison in.
NE5532AP and NE5534AP are the TI part numbers, also still in production.
The NE5532ANG and NE5534ANG are recommended by Doug Self, thus the ones called out in the BOM because he measured the two sources and found slightly better performance of the Onsemi versions. I'd have to look to determine exactly which article he did this comparison in.
Well it appears some like to push components to their limits.
From the servicing point of view this is a very bad choice, especially having serviced electronic equipment for many years. Designers pushing components to their limit have designed bad!
http://www.farnell.com/datasheets/2008138.pdf
This pdf (T.I.) very clearly states when it's read: -
"Absolute maximum ratings max +22v/-22v" - "recommended operating conditions +15v/-15v.
As we known NE5532ANG and NE5534ANG are the Onsemi types, and the NE5532AP and NE5534AP are Texas Instrument types.
In the Elektor's April, May & June's issues, for the Preamplifier 2012, it does states the ANG types, but they are not that easy to get hold of, and major component suppliers (e.g. RS Components, Farnell) don't have them.
Mouser UK list them as obsolete! Not surprising since it's now 2017.
As a constructor and to keep shipping costs down to source 99% of the components from Farnell was ideal.
I'm preferring to base my build as stipulated in the Elektor project and not from any other Tom, Dick or Harry.
From the servicing point of view this is a very bad choice, especially having serviced electronic equipment for many years. Designers pushing components to their limit have designed bad!
http://www.farnell.com/datasheets/2008138.pdf
This pdf (T.I.) very clearly states when it's read: -
"Absolute maximum ratings max +22v/-22v" - "recommended operating conditions +15v/-15v.
As we known NE5532ANG and NE5534ANG are the Onsemi types, and the NE5532AP and NE5534AP are Texas Instrument types.
In the Elektor's April, May & June's issues, for the Preamplifier 2012, it does states the ANG types, but they are not that easy to get hold of, and major component suppliers (e.g. RS Components, Farnell) don't have them.
Mouser UK list them as obsolete! Not surprising since it's now 2017.
As a constructor and to keep shipping costs down to source 99% of the components from Farnell was ideal.
I'm preferring to base my build as stipulated in the Elektor project and not from any other Tom, Dick or Harry.
Last edited:
The other thing is that the data sheets say that the specs/curves are valid for +/-15V. That does not mean that they will be better or worse at different voltages, ONLY that at +/-15V you can rely that they meet the published specs!
So that has nothing to do with reliability and/or whether they actually could be 'better' at different voltages!
You should really actually READ this stuff:
'Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are stress ratings only, and functional operation of the device at these or any other conditions beyond those indicated under Recommended Operating Conditions is not implied. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability. '
It's like an insurance policy. Use +/-15V and we guarantee the shown specs. But that does not mean you can get actually better performance (or worse) at other supplies. Just stay away from the abs. max.
If abs max is +/-22V, then it would be fine to operate them at +/-17V. Only, the published +/-15V specs are no longer guaranteed, but they may be actually better! For one thing, max output levels will be higher.
Jan
So that has nothing to do with reliability and/or whether they actually could be 'better' at different voltages!
You should really actually READ this stuff:
'Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are stress ratings only, and functional operation of the device at these or any other conditions beyond those indicated under Recommended Operating Conditions is not implied. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability. '
It's like an insurance policy. Use +/-15V and we guarantee the shown specs. But that does not mean you can get actually better performance (or worse) at other supplies. Just stay away from the abs. max.
If abs max is +/-22V, then it would be fine to operate them at +/-17V. Only, the published +/-15V specs are no longer guaranteed, but they may be actually better! For one thing, max output levels will be higher.
Jan
Last edited:
I'm finding this discussion interesting as I'm building the Doug Self Linear Audio preamp at the moment and was wondering about the recommendation of running it at +/-17V, I understand it's to maximise SNR but wonder what would be the difference over +/-15V as I have a suitable supply already
What a lot of twaddle.
> to maximise SNR
Higher voltage does NOT reduce hiss. (May slightly increase hiss.)
Higher voltage allows higher SIGNAL.
Op-amps were devised for Analog Computers. Vacuum-tube opamps were worked with 100V maximum signal. Transistor opamps redefined "max" as 10.00V. Because there are a few volts loss in the opamp, this led to the "+/-15V" standard Recommendation. To cover regulator tolerance the AbMax rating was +/-18V. This was gotten with a "40V process" which with normal foundry slop will reliably give breakdown over 40V, giving an extra margin against foundry bad-days.
The "40V process" will usually breakdown nearer 50V. By telling the foundry "good days only!", and spot-checking each wafer, you can VERY reliably rate the chips for 44V. Most opamp users don't do that, so the 36V rating is given for most chips. The '5534/5532 chips have ALWAYS had the 44V number in the specs, the foundry knows that; also the old factory-slop is better controlled than in the 1970s.
The "Suggested Operating Condition" is mostly political/marketing. 99.44% of users do not need large swings, do not need high voltage. +/-15V has been "the" general rating condition for all "40V" opamp chips. They may not have any other power supply handy at the test station.
A large part of '5532 production goes to Studio Audio, where levels inside complex signal chains may not be known or controlled by the studio engineer. High rail voltages cover these peaks. +/-18V is routine. '5532s do not fail at this point. Doug Self worked on (designed and serviced) MANY very large studio consoles.
I've run "36V" chips on 34V and 36V for years. Not as much chip-time as Dough though.
> this thread has lost its way a bit!
Truth from the past indeed.
Do "we" have many-knob complex signal paths?
How high is the signal in a hi-fi preamp? Rarely over 3V peak.
You could run the preamp on +/-9V and never run out of "headroom". (Try it.)
> if you are not happy with that .. then write direct to Elektor.
> If you are not happy with that then write to Farnell!
They don't care to understand the specific project. Elector may know that preamps work every day with +/-15V and don't care to study-up on the '5532. Farnell just sells stuff. They are just quoting the datasheet; you should look-at and complain-to the maker, not Farnell.
Run more than +/-5V, or +/-20V. Even +/-22V if you are SURE it can never hit +/-22.1V.
> to maximise SNR
Higher voltage does NOT reduce hiss. (May slightly increase hiss.)
Higher voltage allows higher SIGNAL.
Op-amps were devised for Analog Computers. Vacuum-tube opamps were worked with 100V maximum signal. Transistor opamps redefined "max" as 10.00V. Because there are a few volts loss in the opamp, this led to the "+/-15V" standard Recommendation. To cover regulator tolerance the AbMax rating was +/-18V. This was gotten with a "40V process" which with normal foundry slop will reliably give breakdown over 40V, giving an extra margin against foundry bad-days.
The "40V process" will usually breakdown nearer 50V. By telling the foundry "good days only!", and spot-checking each wafer, you can VERY reliably rate the chips for 44V. Most opamp users don't do that, so the 36V rating is given for most chips. The '5534/5532 chips have ALWAYS had the 44V number in the specs, the foundry knows that; also the old factory-slop is better controlled than in the 1970s.
The "Suggested Operating Condition" is mostly political/marketing. 99.44% of users do not need large swings, do not need high voltage. +/-15V has been "the" general rating condition for all "40V" opamp chips. They may not have any other power supply handy at the test station.
A large part of '5532 production goes to Studio Audio, where levels inside complex signal chains may not be known or controlled by the studio engineer. High rail voltages cover these peaks. +/-18V is routine. '5532s do not fail at this point. Doug Self worked on (designed and serviced) MANY very large studio consoles.
I've run "36V" chips on 34V and 36V for years. Not as much chip-time as Dough though.
> this thread has lost its way a bit!
Truth from the past indeed.
Do "we" have many-knob complex signal paths?
How high is the signal in a hi-fi preamp? Rarely over 3V peak.
You could run the preamp on +/-9V and never run out of "headroom". (Try it.)
> if you are not happy with that .. then write direct to Elektor.
> If you are not happy with that then write to Farnell!
They don't care to understand the specific project. Elector may know that preamps work every day with +/-15V and don't care to study-up on the '5532. Farnell just sells stuff. They are just quoting the datasheet; you should look-at and complain-to the maker, not Farnell.
Run more than +/-5V, or +/-20V. Even +/-22V if you are SURE it can never hit +/-22.1V.
I will go-back on my hot words.
"Most"(?) people today use "known" sources. DACs can only output so much, hard limit. Tape is soft-limit but there's only so-much can be got from a tape.
Phono-disks..... there's really no limit what the cutter can do for a groove. Good cutters can cut a groove no normal needle can play. This headroom is rarely called upon, partly so the disk will play well, but mainly because wiggle-width and time-per-side are contradictory factors, and disk buyers don't love 13-minute sides.
However a cutter can put a short BIG passage on.
And we now know that ticks and pops can be MUCH higher than expected recorded level.
OTOH some phono-disks are recorded "quiet", so we need extra gain in the system. Since the problem is acute in phono, we often like good phono gain.
The classical computation goes roughly like: we can expect a Stanton/Pickering/Shure to max-out at 70mV clean mid-band. We may enjoy a gain of 100 to bring-up 10mV soft-cut records to "loud". Therefore we aim for 7V RMS or 10V peak. This fits comfortably in a 36V supply, and probably in a 30V supply.
But needles improve, some cutters are pushing the limits, so the old 70mV may not be keeping up with the extreme-hot recordings.
And ticks/pops can be ??? more. Classically we just allow them to clip down-chain. The rise of interest in digitizing, and processing ticks/pops digitally, may mean more concern for recovering those peaks without extra damage (confuses the algorithm or incites added processing damage).
So I could see the PHONO stage working on extra high rails. Indeed the local amplifier runs 60V (+/-30V), though I doubt it ever needed it.
But OTOH: a general rule in audio is that if say 15V makes trouble, 18V or even 22V is not much better. These are only 1.5db and 3dB increases. We know if we clip a 10 Watt loudspeaker amp, we want 6dB to 10dB more power to justify the change.
And also that less-gain is a ton cheaper than more-supply. If your gain of 100 preamp is clipping, shim it to gain-of-50. (Yes, this can compromise simple NFB RIAA, but that cat can be skinned many ways, at least a zillion posted on these forums.)
"Most"(?) people today use "known" sources. DACs can only output so much, hard limit. Tape is soft-limit but there's only so-much can be got from a tape.
Phono-disks..... there's really no limit what the cutter can do for a groove. Good cutters can cut a groove no normal needle can play. This headroom is rarely called upon, partly so the disk will play well, but mainly because wiggle-width and time-per-side are contradictory factors, and disk buyers don't love 13-minute sides.
However a cutter can put a short BIG passage on.
And we now know that ticks and pops can be MUCH higher than expected recorded level.
OTOH some phono-disks are recorded "quiet", so we need extra gain in the system. Since the problem is acute in phono, we often like good phono gain.
The classical computation goes roughly like: we can expect a Stanton/Pickering/Shure to max-out at 70mV clean mid-band. We may enjoy a gain of 100 to bring-up 10mV soft-cut records to "loud". Therefore we aim for 7V RMS or 10V peak. This fits comfortably in a 36V supply, and probably in a 30V supply.
But needles improve, some cutters are pushing the limits, so the old 70mV may not be keeping up with the extreme-hot recordings.
And ticks/pops can be ??? more. Classically we just allow them to clip down-chain. The rise of interest in digitizing, and processing ticks/pops digitally, may mean more concern for recovering those peaks without extra damage (confuses the algorithm or incites added processing damage).
So I could see the PHONO stage working on extra high rails. Indeed the local amplifier runs 60V (+/-30V), though I doubt it ever needed it.
But OTOH: a general rule in audio is that if say 15V makes trouble, 18V or even 22V is not much better. These are only 1.5db and 3dB increases. We know if we clip a 10 Watt loudspeaker amp, we want 6dB to 10dB more power to justify the change.
And also that less-gain is a ton cheaper than more-supply. If your gain of 100 preamp is clipping, shim it to gain-of-50. (Yes, this can compromise simple NFB RIAA, but that cat can be skinned many ways, at least a zillion posted on these forums.)
My Elektor Q-Watt amp (dual ch.) is now up and running, quality is spot on!
Post #368
http://www.diyaudio.com/forums/chip-amps/301100-my-q-watt-project.html
I hope now to return to the Elektor Preamplifier 2012 as soon as possible
Post #368
http://www.diyaudio.com/forums/chip-amps/301100-my-q-watt-project.html
I hope now to return to the Elektor Preamplifier 2012 as soon as possible
I am using the R.S. case 665-7719 | PF-19 Ventilated Rackmount Enclosure with Handles, 2U, 84 HP, 245mm Deep |
Might anyone having used this case have the Front Panel Designer files they used?
With thanks
Might anyone having used this case have the Front Panel Designer files they used?
With thanks
Component query
If i'm to use the Front Panel Designer files from post 18, then based on muskyhuntr's post 6, i would like to ask about the following:-
Power On/Off switch used and order number?
Toggle Switches used for: -
Tone Defeat
MM / MC
IEC Amendment
Mains input socket (i need 240v) used and order number?
With thanks
If i'm to use the Front Panel Designer files from post 18, then based on muskyhuntr's post 6, i would like to ask about the following:-
Power On/Off switch used and order number?
Toggle Switches used for: -
Tone Defeat
MM / MC
IEC Amendment
Mains input socket (i need 240v) used and order number?
With thanks
I have finally been able to get back to this project build after so long.
Can anyone please give me any advise and steps that i should take and be aware of when doing the wiring?
Are there straight forward guides available of all the connections, including the d.c. connections?
With thanks
Calpe
Can anyone please give me any advise and steps that i should take and be aware of when doing the wiring?
Are there straight forward guides available of all the connections, including the d.c. connections?
With thanks
Calpe
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.