1audio said:
We use PhotoMOS opto-couplers all the time, but the MOS output transistor is not a power transistor. They are convenient, but I don't think they are nearly as cheap as BJT opto-couplers.
Cheers,
Bob
<<http://www.hint.no/utdanninger/iu/linker/datablad/IL300.pdf>> Check these out. They compensate for the inherent non linearity of the opto-coupler. I've used them to couple analog sensor outputs where isolation was needed.
anatech said:
if I get what you are saying correctly, the motional mass characteristic alters the sonics in a very audible and derogatory manner.
Mike, those 'linearized' optocouplers look very interesting, and could be very valuable in certain situations.
john curl said:
Hi John,
My audio related interest was as a way to interface my line out on my computer with my preamp for copying records to my hard drive. When I hooked up directly the sound from the main system sounded dirty until I pulled the cables to the PC. With the opto's in place it stayed clean. I haven't done any triple blind testing 😉 , but it makes the transfer of my 30 year old vinyl sound good in my car CD player on my way to work...
They are used in one of my sensor PLC's (Moore Industries) as a way to optically couple analog signals into the AD converters without grounding issues.
Regards, Mike.
Dennis Feucht has a nice design, specs:
high-speed: ³ 100 kHz; 150 kHz -3 dB bandwidth typical
high-linearity: 0.02 % THD at 1 kHz; 0.1 % at 30 kHz, 1 V pk-pk out
large dynamic range with low distortion; < 0.5 % THD for 5 V pk-pk input
250 ppm/°C drift typical
nominal unity gain; independent output gain (scale) and offset adjustments;
300 V isolation minimum, limited only by optocoupler ratings
low manufacturing cost: < $10 parts cost
uses differentially-driven linear optocouplers (LOCs)
unique dual-path feedback topology reduces THD caused by LOCs by x 20
feedback paths from differential LOCs cause error cancellation
I have a copy somewhere but can't find it now. More at www.innovatia.com, Dennis' website.
He also uses the IL300 IIRC.
Jan Didden
high-speed: ³ 100 kHz; 150 kHz -3 dB bandwidth typical
high-linearity: 0.02 % THD at 1 kHz; 0.1 % at 30 kHz, 1 V pk-pk out
large dynamic range with low distortion; < 0.5 % THD for 5 V pk-pk input
250 ppm/°C drift typical
nominal unity gain; independent output gain (scale) and offset adjustments;
300 V isolation minimum, limited only by optocoupler ratings
low manufacturing cost: < $10 parts cost
uses differentially-driven linear optocouplers (LOCs)
unique dual-path feedback topology reduces THD caused by LOCs by x 20
feedback paths from differential LOCs cause error cancellation
I have a copy somewhere but can't find it now. More at www.innovatia.com, Dennis' website.
He also uses the IL300 IIRC.
Jan Didden
janneman said:
Thanks Jan,
The website is not working for me right now. The challenge to working with these is the need for independent supplies for the op-amps on both sides of the couplers. In my application the preamp side was easy and I opted for a 24vdc wall-wart supply and a DC-DC converter to get the +/- 15 for the input side. I didn't want to use the PC supplies, yet I wasn't shooting for high-end results.
Mike.
1audio said:
1audio said:
Hi Demian,
I've been thinking about your description and understand what your doing now. It's definitely a unique approach relative to my experience and from the reviews of the P9 posted on the web it works too. I'm sure, as they say, the devil is in the details.
I was interested in the descriptions of the shielding that I read and it made me wonder if your using a center tapped transformer and where you're connecting the shields into the ground system. In the supply/ control box or in the amplifier box (where the input and outputs connect to the rest of the system components)?
Regards, Mike.
MikeBettinger said:
The supplies are all floating with no connections to anything and the shields are tied to the ground references in the analog chassis at the connector plane. Its not ideal but it works within the constraints of normal audio practice.
janneman said:
janneman
I started this thread 31 january 2003:
Audio Signal Isolation Techniques - Analogue Optocouplers
http://www.diyaudio.com/forums/showthread.php?threadid=10580
In the very first post I attach a short description of TIL300
which is the Texas name for IL300
Attachment:
http://www.diyaudio.com/forums/attachment.php?s=&postid=121254&stamp=1044033991
At the time, soon 6 years ago, wasn't much interest.
But sometimes, sooner or later,
people will catch up and discover good ideas of the possibilities for creative audio designs.
Lineup 🙂 regards to Jan
Hi Lineup,
Yes, interest is there, but if you see what Mr Feucht had to do to linearize those opto-couplers to get a reasonable linear performance over the audio band, that interest will quickly die, I predict! IIRC there are at least 3 optocouplers and 6 opamps. For one channel. Opamp haters will have a field day!😉
Edit: 2 x TL300A, 6 x opamp...
(I paid for the design package, Mr Feucht lives from his designs, so I'm not going to publish a copy here).
Jan Didden
Yes, interest is there, but if you see what Mr Feucht had to do to linearize those opto-couplers to get a reasonable linear performance over the audio band, that interest will quickly die, I predict! IIRC there are at least 3 optocouplers and 6 opamps. For one channel. Opamp haters will have a field day!😉
Edit: 2 x TL300A, 6 x opamp...
(I paid for the design package, Mr Feucht lives from his designs, so I'm not going to publish a copy here).
Jan Didden
I wrote 'linearized' because I am relatively sure that the 'linearization' is relative, and not perfect. Some applications require some sort of isolation to operate at max performance. I suspect that for servos, for example, or bioelectronics, this could be a wonderful solution. I doubt that this would replace a quality transformer for the audio bandwidth, however. The problem is often due to the distortion residual that is left after subtraction. If it is higher order, then the distortion is effectively still going to be audible when using this device.
If there is nothing more to say about 'isolation' we might talk about the differences between the Parasound JC-2 and the CTC Blowtorch. As many may be aware, the Parasound JC-2 has won a fairly respectable position in both 'Stereophile' and 'The Absolute Sound' even though Parasound does not advertise in either publication. The 'heart' of these two preamps were designed by the same CTC design team. However, the design decisions were different, and the cost per feature is very different. It is very much like the difference between a Porsche 944 and a Porsche 911, in design tradeoffs. More later.
JC-2
That would be of great interest.
From the more or less correct BT schematics flying around here compared to JC-2 schematics I got from some of the persons hanging around this thread, it seems the JC-2 uses more transistors (JFETs).
I would like to here more about how and why the Blowtorch sounds better than a more complex design, and why the JC-2 has to be more complex.
Also, if I remember the JC-2 schematics flying around, it has negative feedback as well.
Regds
Rolv-Karsten
john curl said:
That would be of great interest.
From the more or less correct BT schematics flying around here compared to JC-2 schematics I got from some of the persons hanging around this thread, it seems the JC-2 uses more transistors (JFETs).
I would like to here more about how and why the Blowtorch sounds better than a more complex design, and why the JC-2 has to be more complex.
Also, if I remember the JC-2 schematics flying around, it has negative feedback as well.
Regds
Rolv-Karsten
okay.
😎 so you post at least a conceptual diagram of them both 😎
you know, without any component values .. even without transistor names, if you like
It is not very easy for the general public to talk about what only the few knows
If you want to talk we can, unless you make it impossible
and you like to keep discussing above our heads
with those few
those that 'knows it all' ... more than me & most diyaudio members at this forum
Lineup 🙂 thinks maybe Mr Curl had in mind we all should
rush to buy each our own Blowtorch and Parasound JC-2
in order to be able to discuss the differences
John,
can you give me an advice, please.
I want to find one N-JFET that has got VGS 2-4 Volt for like 1-5 mA
and that will work reasonably well at VDS >= 20-25 Volt.
Is for input stage.
I have used 2N3819 (rated 25 VDS) at VGS 2.20V and 0.80mA.
But you have any other suggestion?
thanks 🙂
Lineup
can you give me an advice, please.
I want to find one N-JFET that has got VGS 2-4 Volt for like 1-5 mA
and that will work reasonably well at VDS >= 20-25 Volt.
Is for input stage.
I have used 2N3819 (rated 25 VDS) at VGS 2.20V and 0.80mA.
But you have any other suggestion?
thanks 🙂
Lineup
- Status
- Not open for further replies.