If you take a real amplifier circuit and substitute all the transistors and providing they are suitably rated for voltage and current then generally the circuit will still work as before. Same for simulation, swap a BC546 for a 2N3704 for a BFY51 for 2N5551 and it will all work and sim essentially the same.
Try this one. Look at the models. All picked totally at random from devices that generally meet the spec for a circuit like this.
Try this one. Look at the models. All picked totally at random from devices that generally meet the spec for a circuit like this.
Thank you so much again This shows how the actual topology and voltage and bias points are important Great
But still for a same duty some bjts should be more performant than other (i am guessing here)
My main problem now is basic ignorance I guess there must be an optimal working point for a bjt in terms of Vceo and Ic
I say this because they are the parameters most highlighted in the datasheets
The nest goal is to understand from the values that come out from the simulation if the various bjts are working optimally or not
This is when passion is facing ignorance
But still for a same duty some bjts should be more performant than other (i am guessing here)
My main problem now is basic ignorance I guess there must be an optimal working point for a bjt in terms of Vceo and Ic
I say this because they are the parameters most highlighted in the datasheets
The nest goal is to understand from the values that come out from the simulation if the various bjts are working optimally or not
This is when passion is facing ignorance
Audio in the grand scheme of things is very non demanding, we have a very low bandwidth and frequency range to deal with. If you were designing an RF stage or a VHF or UHF or S Band tuner stage it is a very very different matter indeed.
With audio things such as small signal current gain or hFE is important for say front ends to get the very best performance and low junction capacitance might be important for some aspects of a design but often not as much as you might think. Choose an inappropriate device for the RF design and it simply won't work at all, there the characteristics are of paramount importance.
Look at the sim in Examples called 'UHF Preamp'. That might be a circuit for a TV aerial booster.
Look at the gain in db at the left:
Now we swap the only transistor for a high gain device that we might use in audio. The performance now is worse than useless and it is attenuating the signal as the transistor is hopeless at these frequencies:
Look at the gain in db at the left:
Now we swap the only transistor for a high gain device that we might use in audio. The performance now is worse than useless and it is attenuating the signal as the transistor is hopeless at these frequencies:
Maybe But opamps aside and speaking of discrete very few design can achieve THD -120dB
I am checking but most of them stops at -100dB I agree that that -20dB difference maybe is not audible ... still
dear Mr Mooly I have to be honest
Many times the most ambitious ones are also the least gifted, the most ignorant and stupid and I am part of the group with honor
I had made some years ago a circuit with an operational am the infamous ad797 masterpiece after reading an excited article on The Audio Critic magazine.
I had to try it. I use a pcb from a kit and a LM317/337 +/-15VDC decent kit.
The potentiometer made noises at the start of the rotation and so I thought I were doing something wrong.
Later I discovered that a simple 100 ohm resistor would have isolated the wiper from the ground and cured the problem. (After that experience i have always placed a 100 ohm at the pot exit then a cap and i solder the wiper to the cap No more issues with pot noise)
What really excites me is the idea of being able to put together a very minimalist discrete buffer with THD at -120dB, with two or max three active components and some resistance and capacitance around it.
Let's say I'm trying, that's why I have the clear feeling that this is possible.
If anything, it will take the best BJTs or MOSFETs, higher voltages, but in my opinion it can be done.
After that i could pass to power amps with the convinction to have a great volume controller already
Thank you very much indeed I had this feeling that starting well is very decisive for the final result
In the past I have heard the same power amp with different preamps. The sound changed in an impressive way
Since that experience i have been fascinated by the line preamps than the power amps
I do not listen to LPs at all The set up can be a nightmare
The AD797 is a very high performance opamp and is very demanding in terms of the board layout and the circuit application. It sounds very much like your project might have been oscillating as the pot wiper reached ground. Adding a 100 ohm isn't a normal thing to do with most designs.
The best opamps are very difficult to even measure at those sort of levels but there are tricks that can be used to make the distortion 'bigger' and so be able to calculate what the real distortion is.
Many opamps use this method:
I first saw that done a long time ago... its clever but I think more curiosity value
as much as anything.I still think you should build some simple circuits yourself and for you to test some opamps and see what you think... don't believe all you read... you make your own mind up.
Have a look at post #68 here. Did you know you can actually 'listen' to your simulations?
https://www.diyaudio.com/community/...in-the-pearl-3-phonostage.410071/post-7631271
That was just a bit of fun but you can feed an audio signal as an input to the sim (a .wav file) and save the result as a .wav
The attached file in that post is a recording of the simulation, two channels in stereo.
It is very likely I remember i follow a schema shown in the datasheet Vgain was 2 ? with a small cap across the feedback resistor
Aside the very bad noise at the start then it was sounding quite clean But i ditched everything I understood that opamps can be nasty when used badly
Yes A friend of mine built a Bride of Zen line stage Then he passed to tubes and left me the BOZ Instead of using the pot at the output i decided to order a 20klog series attenuator made with Holco resistors When i mounted it the same exact problem at zero level
The seller advised me to put a small resistor in series with the wiper and it worked very well Complete silence
Now i put a 100 ohm resistor everywhere
But i agree that is not normal ... maybe there must be some kind of noise circulating in the groung i do not have a clue
I am feeling naive to want to stick with discretes But everyone here propose discrete circuitsThe best opamps are very difficult to even measure at those sort of levels but there are tricks that can be used to make the distortion 'bigger' and so be able to calculate what the real distortion is.
I first saw that done a long time ago... its clever but I think more curiosity value
I still think you should build some simple circuits yourself and for you to test some opamps and see what you think... don't believe all you read... you make your own mind up.
I wonder why if almost never they can provide the same performance of simpler opamps This confuses me
this is more than great ! i cannot believe it really This SW is really amazing Going there immediately !!!amazing This must be AI ...
Ok lets say that i am now set on the 2604 to be used as unity gain buffer after looking at its datasheet
I would like to buy a pcb with all the traces to host a dual opamp The dual power supply just to start could be a simple ebay kit i am afraid
I tried the breadboard but i much prefer a pcb
Can i use one of those intended for kit using the NE5532 ? i guess they can be found online quite easily ?
Attachments
Last edited:
The OPA2604 was a wonderful device but its out of production now and you can pretty much guarantee that any you see for sale are fake. I picked that one because the data sheet explained the 'Noise Gain' method well.
I was wondering about an eBay special
Why not try a tone control stage that uses a couple of opamps (and I wouldn't be surprised if the the opamps were fake) and once you have it working try some different types that you have bought from authorised suppliers. I think you should try the worst vs some highly thought of opamps and that way you get an idea of what you are listening for.These kind of things:
Hi good morning I saw them but i guess that the ones offered to use with NE5534/5532 can be more basic
Besides i have hated tone controls since childhood
I have a question more out of curiosity
I have here the best line stage made by Rotel (the schema i mean)
I am stuck with a MAT27W-A part It looks like two diodes one above the other ???
Is there something electrically equivalent from a sim point of view ?
As i said it was used in the TOTL Rotel line preamp of the Michi series (Model: RHA-10; Model range: Michi; Type: Active preamplifier; Years of manufacture: 1991 - 1997; Made in: Japan)
Besides i have hated tone controls since childhood
I have a question more out of curiosity
I have here the best line stage made by Rotel (the schema i mean)
I am stuck with a MAT27W-A part It looks like two diodes one above the other ???
Is there something electrically equivalent from a sim point of view ?
As i said it was used in the TOTL Rotel line preamp of the Michi series (Model: RHA-10; Model range: Michi; Type: Active preamplifier; Years of manufacture: 1991 - 1997; Made in: Japan)
That part is essentially just a 'double diode' of two series connected silicon diodes like IN418. They are used for biasing output stages. When you sim any circuit it is very very important to make sure the final bias conditions in the output stage are correct. The product might run at say 30ma flowing in the output transistors and if you sim the circuit you might find either low, no or even high bias and that is because the forward turn on voltage of the actual chosen parts and the models you are using vary a little. It can make a massive difference to the sim (remember the 741 that seems to have zero bias and the FFT).
So always tweak the bias to get the current correct.
You can also often substitute the diode/s in the sim for a voltage source and set the value to say 1.1 volts to simulate two series diodes and the volt drop at very low current.
Cut the transistor out and replace with a voltage source to equal the voltage developed across the original.
So always tweak the bias to get the current correct.
You can also often substitute the diode/s in the sim for a voltage source and set the value to say 1.1 volts to simulate two series diodes and the volt drop at very low current.
Cut the transistor out and replace with a voltage source to equal the voltage developed across the original.
Dear Karl and guys. Is there any simple method to simulate an electret mic with its internal FET?. This is for a ham SSB transmitter and to round the complete gain and set bandpass to 300 to 3000hz, including the RF rejection filter at the input of it. Naturally it isn't a hifi device. And I need to estimate the gain necessary as to not saturate the ring modulator for SSB.
I have LT IV in a Win XP machine.
Many thanks in advance.
method to simulate an electret mic with its internal FET?. This is for a ham SSB transmitter and to round the complete gain and set bandpass to 300 to 3000hz, including the RF rejection filter at the input of it. Naturally it isn't a hifi device. And I need to estimate the gain necessary as to not saturate the ring modulator for SSB.I have LT IV in a Win XP machine.
Many thanks in advance.
On the reference Yahoo site for LTspice, there were a number of electrets available:
https://www.diyaudio.com/community/...ectret-microphone-ltspice.270047/post-4231442
Now that it migrated to "io" everything is upside down, but the content is probably still there, somewhere... With some patience and determination (and maybe the wayback machine), it should be possible to locate it in the new site
https://www.diyaudio.com/community/...ectret-microphone-ltspice.270047/post-4231442
Now that it migrated to "io" everything is upside down, but the content is probably still there, somewhere... With some patience and determination (and maybe the wayback machine), it should be possible to locate it in the new site
I've nothing like that set up but I can't see why you can not just set up a voltage source as the capsule and set the output to the correct level. You can also add the required impedance to the settings.Dear Karl and guys. Is there any simple
https://sound-au.com/articles/mic-electret.htm
I agree with Mooly, it's a "simple" voltage source with maybe a 1k source impedance representing the resistor pull-up that electret mic elements use. The Rod Elliot article mentions distortion at high volume level (which I imagine can happen with a regular voice one inch from the mouth). The "Linkwitz mod" fixes this for the no-longer-available Panasonic WM-61A, and I wonder if it could be done with similar electret elements. See the .gif link in the sentence "Look at the microph1.gif page from my notebook for further details" on this page:
https://www.linkwitzlab.com/sys_test.htm
https://www.linkwitzlab.com/sys_test.htm