My sim shows an almost perfect fit... I would only use 6k8 // 1M4 instead of 6k8 // 2M2.
The output impedance of the cascode beeing ~ 2k2, it adds to the 47k in the riaa... very good.
Very slim reactions, see graphs.
Attachments
Hey folks!
Could somebody tell me the most recent version of the Salas RIAA pdf please? Is it 11.02.F0? (what do the numbers correspond to by the way?).
Thanks,
John.
Could somebody tell me the most recent version of the Salas RIAA pdf please? Is it 11.02.F0? (what do the numbers correspond to by the way?).
Thanks,
John.
Yes that is the one. New stuff discussed is some beta modding to see if we can positively revise. Still to do that successfully to a widely liked design it will take time and it may very well be a thumbs down in the end. Numbers mean Feb. 2011 and document code, to allow correct reference.
Thanks Salas! 🙂
Perhaps we could keep the most current link to the PDF in your first post in this thread?
J.
Perhaps we could keep the most current link to the PDF in your first post in this thread?
J.
It's linked there already John. If any official revision will come up, it will be linked there too.
Your original sketch jpg is in the first post, but no pdf link that I can see?
Edit: ah ok it's in the highlighted txt above - maybe change the link title to 'latest build PDF' just to make it obvious, to idiots like me 😉
Edit: ah ok it's in the highlighted txt above - maybe change the link title to 'latest build PDF' just to make it obvious, to idiots like me 😉
Last edited:
"PDF instructions/Update" I renamed it.
You're a good man Salas - thanks 😎
I've been working on my RIAA boards doing some tidying. I need to get it in some form of box.
I used these nice little through-hole eyelets to be able to easily change parts and make connections.
I just found out they are steel. See them here in my earlier post.
I realize this sounds naive, but is this not good practice? I have quite a few of these eyelets in the circuit path.
Regards- Kent
Hey Kent 🙂
Your build is one I've been following with great interest since ploughing (backwards!) through this thread all afternoon!
Firstly, would you mind if I emulated your very neat layout please?
Secondly, how did your stage come out sounding in the end? Would love to hear your views.
Cheers.
John.
Very slim reactions, see graphs.
I notice a nitid deviation on the second inflection point between second and thir graph.
K369 for noise is as using 2 K170 roughly. Yfs and Ciss go up together, so no free lunch, it shows some advantage in lower frequencies for a special kind of noise named generation recombination as Scott W. explained. Here you are, another toy attached.🙂
This will be my next build.
I like the sound of my actual 48d "stock" version so much, I decided not to tamper with it and use it as "pivot".
I will do this 57db as soon as I collect enough CHF to buy the parts needed 🙂
Really happy about this.
I notice a nitid deviation on the second inflection point between second and thir graph.
I did not get that, can you explain in more words?
Fourth graph presents an inflection point near 3kHz.. tirth graph presents the same inflection at a much higher freq (5 to 6kHz)
The scale is so much zoomed in that with real parts tolerances I am not sure that we will ever be able to emulate those curves, the idea was to initially have a very faint droop as we progress in higher frequencies. Still if the modded gain blocks will work to satisfaction, those curves will possibly need to be more heavily voiced to subjective synergy or left alone. Till then, some road. But its a non problematic starting point, not too deviant IMHO. What do you think?
I believe the stock solution (6k8 // 2m2) will give sound a bit more "attack" but nothing to worry about... The other values are very good indeed.
This will be my next build for sure 🙂
This will be my next build for sure 🙂
Hey Kent 🙂
Your build is one I've been following with great interest since ploughing (backwards!) through this thread all afternoon!
Firstly, would you mind if I emulated your very neat layout please?
Secondly, how did your stage come out sounding in the end? Would love to hear your views.
Cheers.
John.
Hi John- Please feel free to emulate away!
How does it sound? Great! I only have a Mitsubishi Integra preamp to compare it to. My Salas RIAA build has more dynamics and life. Very happy with it's sound. Of course there are many variables and subjectivity. I'm also using a lab supply to power the shunts as I've been too distracted/busy to build a transformer/CLC power supply.
It doesn't have quite as much gain as the Integra but plenty loud enough. Possibly the latest ideas on this thread will address that. Still good to go. I hope I can get into in a box for Burning Amp in October! 😀
-Kent
58db cascoded build
Hi Salas
For Q8, can I use a BF245C instead of a BF245 A ?
It sets the current in the leds that provide bias for Q3 (folded cascode trannie).. do you believe higher current might modify bias too much or is it about noise ?
In alternative, I can use a suitable 6mA K170 right ?
Hi Salas
For Q8, can I use a BF245C instead of a BF245 A ?
It sets the current in the leds that provide bias for Q3 (folded cascode trannie).. do you believe higher current might modify bias too much or is it about noise ?
In alternative, I can use a suitable 6mA K170 right ?
Attachments
You may either put a resistor between G&S BF245C pins with a value that will get it down to circa 5mA (try 470R) and take current output from G pin, or use 2N5457 3-5mA natural. My BF245A little stash averages at 3.5mA for instance. They are both low capacitance higher RDS ON JFETs but suitably quiet also that will meet the 4 Leds impedance better for minimal harmonic noise pick up of CCSed Leds Vref in that cascode transistor's base bias system. Not a big deal but prefer them to K170 for the job if able.
- Home
- Source & Line
- Analogue Source
- Simplistic NJFET RIAA