In I/V conversions an uncompensated NE5534 (at seemingly < $1) can provide excellent results in a network that supports it. Nevertheless it could be improved upon by an AD827 or other FET input devices if I can find one without paying > $40. The reason I am interested is that I prefer bipolar input devices and wonder if a device like the AD827 is superior because it has a strong reputation in the DIY community.
There is almost no new tube successor
But plenty of newer and better chips at a lower cost
Any reason to use $ chips with old tech?
But plenty of newer and better chips at a lower cost
Any reason to use $ chips with old tech?
Yes, but be kind, let him ask and us to share knowledge.
It's actual as never before, but overpriced due to its DC laser-trimmed precision at each crystal. But, yes, there are an even better uncompensated one - OPA637, especially if you can cope such a beast.
Ah, yes, AD847/27 are well-known performers.
LM4562, LM6171/72, AD8065/66 and even AD844 could be an option too. All known for its good sounding.
Don't care much about cost, or tech (old or new). There is old tech trash and new tech trash. The NE5534 was designed in 1975, almost half century ago as a TDA1034. It is also the device most often sold in lieu of legitimate AD797's, LT1028's, etc., coming from China. The simple test is to look at the trimming.
Tried most of the ones above. Currently settled on the AD844 feeding an NE5534 for I/V. The question is always, can something else be implemented as a whole better? The chip is only partly at issue.
Bought an AD8066 from Ebay/China recently. Powered it up in a +/-15V circuit and it fried. I forgot that it was a 24V part so I reckon it was authentic! Like testing a match. LoL....
Watch out for pretty bad common mode distortion with the 5534/32 when used non-inverting and/or as a follower + large source Rs like >5Kohms. i.e 10K, 47K. Inverting mode is OK. But when IN- must chase IN+ it's quite flawed. FETs have a similar behavior in common-mode but was fixed by Burr-Brown with their Difet process a la the '627 and the earlier '2107. The latter is also very nice sounding, BTW.
A fellow I worked with (really not me...) was testing fuses for continuity. He found then all defective. They were of very low current values and his meter was blowing them... like testing matches.
That is what I have found as well, hence I am using the NE5534 in the inverting mode. The same holds true for the AD844. From the data sheet the AD627 looks like a cascode followed by an inverting cascode, seemingly to support the non-inverting mode on the input. This is one of the reasons I would like to try this.
You won't find the magic in the schematic. It has to do with the Gate-to-Substrate capacitance of the input FETs which varies with voltage. See attached article by TI. TI's name for the fix is Dielectric Isolation or "DI". See the OPA1642 example of stabilized Cgss as in voltage insensitive. Preceding this (c. 1989) was what BB called Difet. Same thing. Did you look at the OPA2107 Difet. It's less $$, is a dual and still can be had in DIP8 if that's a + for you. By the way the OPA/BB627 was always expensive. I believe back in 2009 is was ~50x the price of a 5532 per part = 100x per channel.
Light reading:
Douglas Self – “Op amps in Small-signal Audio Design” EE Times
Part 1
https://www.eetimes.com/op-amps-in-small-signal-audio-design-part-1-op-amp-history-properties/
Part 2
https://www.eetimes.com/op-amps-in-...?utm_source=eetimes&utm_medium=relatedcontent
Part 3
https://www.eetimes.com/op-amps-in-...?utm_source=eetimes&utm_medium=relatedcontent
Part 4
https://www.eetimes.com/op-amps-in-...?utm_source=eetimes&utm_medium=relatedcontent
Douglas Self – “Op amps in Small-signal Audio Design” EE Times
Part 1
https://www.eetimes.com/op-amps-in-small-signal-audio-design-part-1-op-amp-history-properties/
Part 2
https://www.eetimes.com/op-amps-in-...?utm_source=eetimes&utm_medium=relatedcontent
Part 3
https://www.eetimes.com/op-amps-in-...?utm_source=eetimes&utm_medium=relatedcontent
Part 4
https://www.eetimes.com/op-amps-in-...?utm_source=eetimes&utm_medium=relatedcontent
What about the OPA27. Is there an easy measurement that can be done to determine if the part is genuine?
If it's a good/useful part, it's a good/useful part; how long it has been around is not really relevant in that context.
Its VERY expensive manufacturing process gave it significant advantages over other parts. Cheaper processes eventually began to shrink some of the advantage gaps. But these expensive parts are still being made and sold. So, they still have something the cheaper parts don't yet provide. I'll give you a hint that the audio world is by far NOT the biggest user of very high precision op amps ...
The OPA627 is a unity gain stable op-amp with 100dB open loop gain at 20kHz, 60dB at 1MHz. Not a bad GBP for a f***ing old device (not to be confused with the f***ing old proclivities seeming of little concern to older persons... particularly those hard of hearing).
Feeding in a small 1 MHz signal ought to get a gain of 1000x open loop, that if set up alternatively for a gain of 100 ought to result in close to achieving that amplification, as seeming doubtful for fakes to replicate easily.
Feeding in a small 1 MHz signal ought to get a gain of 1000x open loop, that if set up alternatively for a gain of 100 ought to result in close to achieving that amplification, as seeming doubtful for fakes to replicate easily.
I don't see that on the datasheet. Instead, what I see is 57 dB open loop gain at 20kHz and 23dB at 1MHz. A difference of ~ 40dB (100X)
_
