Sonny, PPL already mentioned that he used AD8610 and AD825 opamps with the original adapter without problems. You imply that the adapter is somehow unfit for higher bandwidth devices than the OPA132, which is simply not the case, vis a vis empirical evidence.
I do not believe you can demonstrate an audio amp circuit that works perfectly with bypass caps wired to the adapter pins that oscillates when the caps are wired to traces next to the DIP socket instead. As I mentioned before, the distance on the LC Audio module from the top opamp to the bypass caps is greater than that.
Eeek, you live in Denmark, I hope you don't work for LC Audio, hee hee!
I do not believe you can demonstrate an audio amp circuit that works perfectly with bypass caps wired to the adapter pins that oscillates when the caps are wired to traces next to the DIP socket instead. As I mentioned before, the distance on the LC Audio module from the top opamp to the bypass caps is greater than that.
Eeek, you live in Denmark, I hope you don't work for LC Audio, hee hee!
morsel said:Sonny, PPL already mentioned that he used AD8610 and AD825 opamps with the original adapter without problems. You imply that the adapter is somehow unfit for higher bandwidth devices than the OPA132, which is simply not the case, vis a vis empirical evidence.
I do not believe you can demonstrate an audio amp circuit that works perfectly with bypass caps wired to the adapter pins that oscillates when the caps are wired to traces next to the DIP socket instead. As I mentioned before, the distance on the LC Audio module from the top opamp to the bypass caps is greater than that.
I am happy to tell you that i am not working for them (pyhhh!!).
For their modules.. Their way of bypassing the opamp in general ... It is not my style at all.. Also to thin tracks!
For the LCLOCK with low jitter.. I can tell you that if you measure on the clock signal and twist/push/change the length of the wires for the clocksignal you will see an improvement/degration on the clocksignal.. This is less than perfect!!!
This is not the way to do this.. I would prefer a better cable or even better a seperate DAC unit where the LCLOCK is embedded into to the PCB so that the CLOCK has a stable transmission line/line impedance...The LCLOCK it self is okay..
A lot of people will disagree with me right now!
Maybe you are right about my judging... ;D I am really sorry and i think all should listen more to PPL and you MORSEL right now! than me
By the way i have seen the other modules you have made? You have added a ground plane... Thumbs UP! Not many do this
I design PCB 5 - 7 times a year..
The reason why it sound like i am not happy about this and all the other modules i see has something to do with the EMC test i run on a daily basis.
And i see nothing but trouble when running a EMC test when the enclosure is not done right and the tracks from bypass caps is thin and long for any device.. CPU,Ram,OPAMP,DAC,ADC etc..
And this i can demonstrate!!
I also think when you can se a difference on track thickness and length on digital working at only 5 - 16MHz. Why should this not be the same for an opamp who has a bandwidth of 40MHz and higher?
You can also measure on the supply pins what frequency the opamp is oscillating at when the decoupling is not at its optimum.. But okay i am detail freak.
Again if you modules work perfectly like PPL says, then i take my words back.. Seriously!!!!
Sonny
IC adaptors
Since I seem to have stired things up, Using an adaptor 100Mhz and slower op amps is fine. I have done work with video op omps and favorites like the AD811 and AD846. These are fairly twichy about layout and decoupling. Also stray capacitance at the input pins is a big influence on perfomance. I had a digital data interface op amp circuit that sounded better if bent of the inputs pins up off the board and mounted the feedback resistor on top of the op amp. This was on a VERY carefully done layout. The difference was about 2pF less capacitance. Be very careful with high speed op amps and read the application notes very closely.
H.H.
Since I seem to have stired things up, Using an adaptor 100Mhz and slower op amps is fine. I have done work with video op omps and favorites like the AD811 and AD846. These are fairly twichy about layout and decoupling. Also stray capacitance at the input pins is a big influence on perfomance. I had a digital data interface op amp circuit that sounded better if bent of the inputs pins up off the board and mounted the feedback resistor on top of the op amp. This was on a VERY carefully done layout. The difference was about 2pF less capacitance. Be very careful with high speed op amps and read the application notes very closely.
H.H.
Sonny, there is a significant difference between a digital circuit that is operating and requiring signal integrity at 16MHz .vs. an analog audio circuit operating under 100KHz that happens to use an opamp vulnerable to RF oscillations in the MHz range, in fact, its totally different, apples and oranges. Regardless of 4, 40, or 400MHz opamp bandwidth, it does not change the fact that the purpose of the circuit is AF amplification. It is far easier to shunt a bit of RF to ground than to design a circuit to handle RF signals with integrity. Regardless, the distance from the SO8 through the adapter to the DIP pin is so short it is not going to make much difference one way or the other, even for MHz range applications, as long as the socket is decent quality and the bypass caps are near the DIP socket.
No need to quote everything in bold text, I can see the previous messages just fine.
No need to quote everything in bold text, I can see the previous messages just fine.
Oopss.... bold text .. I am sorry!! It was not ment this way... To make you angry.
Still i do not agree with you. Most new micro's incorporate a PLL wich have to be treated the same way as a opamp.
And i see no difference in i a MHz digital unit or a MHz opamp. They need the same low impedance decoupling. Check applicationnote from .. lets say AD.
Sonny
Still i do not agree with you. Most new micro's incorporate a PLL wich have to be treated the same way as a opamp.
And i see no difference in i a MHz digital unit or a MHz opamp. They need the same low impedance decoupling. Check applicationnote from .. lets say AD.
Sonny
High speed op amps
Poor Sonny. He is getting beat up for what I started. I guess I will stick my 2 cents in:
"there is a significant difference between a digital circuit that is operating and requiring signal integrity at 16MHz .vs. an analog audio circuit operating under 100KHz that happens to use an opamp vulnerable to RF oscillations in the MHz range, in fact, its totally different, apples and oranges. "
Since much of audio is digital audio nowadays this is not as cut and dried as one might think. Most of the study of signal integrity in the digital world is based on analog theory from RF circuits. Also, we live in world saturated with RF noise and EMI. GHz prorocessors, cell phones, microwave, and radio signals are everywhere. RF will get in your circuits. Especially in audio circuits with no input filters and unsheilded interconnect. The diodes in the raw power supply will even provide RF to your "100KHz circuit". I will have more to say on this in the future I am quite sure since I have worked with signal integrity and EMI complience for a living. Further discourse Morsel?
H.H.
Poor Sonny. He is getting beat up for what I started. I guess I will stick my 2 cents in:
"there is a significant difference between a digital circuit that is operating and requiring signal integrity at 16MHz .vs. an analog audio circuit operating under 100KHz that happens to use an opamp vulnerable to RF oscillations in the MHz range, in fact, its totally different, apples and oranges. "
Since much of audio is digital audio nowadays this is not as cut and dried as one might think. Most of the study of signal integrity in the digital world is based on analog theory from RF circuits. Also, we live in world saturated with RF noise and EMI. GHz prorocessors, cell phones, microwave, and radio signals are everywhere. RF will get in your circuits. Especially in audio circuits with no input filters and unsheilded interconnect. The diodes in the raw power supply will even provide RF to your "100KHz circuit". I will have more to say on this in the future I am quite sure since I have worked with signal integrity and EMI complience for a living. Further discourse Morsel?
H.H.
Boff! Bam! Pow!
One of the points I was trying to make is that the analog opamp circuit exists for the AF signal. RF is neither desired nor needed. It is kept out by various means, including bypass caps and even caps in the feedback loop of the opamp, if need be.
The 16MHz digital circuit exists for an RF signal, not an AF signal. You can't just filter spurious RF with a catch all LPF, or you will damage the digital signal. Thus you have to be much more careful with the design.
RFI is a valid concern. I was defending the usability of the opamp adapter with fast opamps, but we have gotten way off topic. Let's talk about good opamps instead.
One of the points I was trying to make is that the analog opamp circuit exists for the AF signal. RF is neither desired nor needed. It is kept out by various means, including bypass caps and even caps in the feedback loop of the opamp, if need be.
The 16MHz digital circuit exists for an RF signal, not an AF signal. You can't just filter spurious RF with a catch all LPF, or you will damage the digital signal. Thus you have to be much more careful with the design.
RFI is a valid concern. I was defending the usability of the opamp adapter with fast opamps, but we have gotten way off topic. Let's talk about good opamps instead.
I tried to pull a spec sheet on the 846 but Analog says it has been discontinued. I took a quick peek at the 811, but I'm running out the door shortly so a thorough read will have to wait. I am more up on the FET input opamps like the 823 and 825. Various people have been saying good things about the (bipolar) 826 recently. Icceman on Headwize got a bunch of different opamps, with more on the way, and he is trying them out in his X-Cans for comparison. Should be interesting reading when all is said and done. Seeya tomorrow.
Good op-amps
Hi, nice to be here-
the newer model of the Lm6172
you might refer to I supsect is lm6181. No idea if there is a dual type (lm6182)
I´ve been recommended the lm6172 for bitstream dac´s analog output stages but the lm6181 for analog circuits
enough talking ´bout thing untried and unseen units for now;-)
Philipp
Hi, nice to be here-
the newer model of the Lm6172
you might refer to I supsect is lm6181. No idea if there is a dual type (lm6182)
I´ve been recommended the lm6172 for bitstream dac´s analog output stages but the lm6181 for analog circuits
enough talking ´bout thing untried and unseen units for now;-)
Philipp
KevinLee said:
Perhaps you should have a look at Thorsten Loesch's article on an extensive Marants mod:
http://www.tnt-audio.com/clinica/cd67.html
Carlos
Carlos,
Thanks for the link, I have know of that article for a couple years and have completed most of the mods listed in the article.
As far as LM6172 goes, the author says "a bit more care must be used in the power supply when using the LM6172". He does not expand upon things any further than that.
I have tried AD827 & OPA2132 in the cd player so far. I found the OPA2132 a little slow for my liking.
I would like to try the LM6172, but I would like to know if I need to add any other parts at the same time to make the op-amp work optimally.
Thanks
KevinLee
Thanks for the link, I have know of that article for a couple years and have completed most of the mods listed in the article.
As far as LM6172 goes, the author says "a bit more care must be used in the power supply when using the LM6172". He does not expand upon things any further than that.
I have tried AD827 & OPA2132 in the cd player so far. I found the OPA2132 a little slow for my liking.
I would like to try the LM6172, but I would like to know if I need to add any other parts at the same time to make the op-amp work optimally.
Thanks
KevinLee
KevinLee said:
What you have to be careful with are two things:
1) Be careful if you are going to use IC sockets (Thorsten recommends not to use sockets).
2) Solder a small size .47uF film capacitor between the V+ and V- legs, directly on the pins.
There's another article you should read on upgrading ICs:
http://www.zero-distortion.com/tests/opamps/opamps.htm
The AD826 is very recommended on it too.
Carlos
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.