Gyula,
I'm sure we will hear from you before then - how can you stay away – happy studying!
Jaka,
I simulated the filter last night and found the "Notch" filter. It VERY sensitive to component values, so I will probably replace L2 with an adjustable inductor to fine-tune the Notch. Simulations show -120dB attenuation at the notch Frequency - will be interesting to compare to real hardware results
Once again many thanks for the Info & Scans.
I'm sure we will hear from you before then - how can you stay away – happy studying!
Jaka,
I simulated the filter last night and found the "Notch" filter. It VERY sensitive to component values, so I will probably replace L2 with an adjustable inductor to fine-tune the Notch. Simulations show -120dB attenuation at the notch Frequency - will be interesting to compare to real hardware results
Once again many thanks for the Info & Scans.
I created a monster
John and Jaka, thanks for the comments.
This is getting a huge project. By building it modulair and in small steps it must be feasable.
John, what do you mean by your remark about the 6 analogue inputs? I'm aware of the dac / adc loop between the dvd and the preamp, but is there an alternative to get the decoded signal in the preamp?? (Without modding several DVD players to get the I2s?)
As for the low jitter clock in the preamp, I distributed the signal throughout the the design when in analogue mode. Easiest for keeping the machinery in synch. I guess.
Now working on the schematic of the AES receiver, using the TI DIR 1703. This schematic is soon to be posted as well as the almost finished H-bridge power supply prototype. No measurments taken yet. Scope is soon to arrive and then I know more.
Gyula, hope to hear from you soon, altough your Class-I is not closely studied by me, due to lack of time, but impressive that it beats the quad and that you made it sound High-end.
Cheers!
John and Jaka, thanks for the comments.
This is getting a huge project. By building it modulair and in small steps it must be feasable.
John, what do you mean by your remark about the 6 analogue inputs? I'm aware of the dac / adc loop between the dvd and the preamp, but is there an alternative to get the decoded signal in the preamp?? (Without modding several DVD players to get the I2s?)
As for the low jitter clock in the preamp, I distributed the signal throughout the the design when in analogue mode. Easiest for keeping the machinery in synch. I guess.
Now working on the schematic of the AES receiver, using the TI DIR 1703. This schematic is soon to be posted as well as the almost finished H-bridge power supply prototype. No measurments taken yet. Scope is soon to arrive and then I know more.
Gyula, hope to hear from you soon, altough your Class-I is not closely studied by me, due to lack of time, but impressive that it beats the quad and that you made it sound High-end.
Cheers!
Attachments
I once achieved a carrier suppression of 83 dB by using a Cauer filter. You won't see anything of the carrier on a scope when a signal of reasonable level is present.
In practive a carrier suppression of 40 dB is sufficient IMO. Just keep in mind that there won't be much current flowing through your speakers at these frequencies, due to their inherent inductance.
You won't radiate much RF at the carrier frequency either, since you need quite a large antenna at these frequencies to get anything up in the air. Things look different at the carrier's harmonics however and it isn't a bad idea to keep an eye on these.
Regards
Charles
In practive a carrier suppression of 40 dB is sufficient IMO. Just keep in mind that there won't be much current flowing through your speakers at these frequencies, due to their inherent inductance.
You won't radiate much RF at the carrier frequency either, since you need quite a large antenna at these frequencies to get anything up in the air. Things look different at the carrier's harmonics however and it isn't a bad idea to keep an eye on these.
Regards
Charles
I found the suggested handbook "Ferrites and Accessories" in English language(!) on a Siemens CD from 1999. With detailed ferrite core and material data and cookbook-like calculations regarding transormers, chokes and resonant inductors. In pdf-format ca. 35MB, 160+files.
Anyone interested?
Jaka, your article is very informative, thanks a lot!
Regards, Timo
Anyone interested?
Jaka, your article is very informative, thanks a lot!
Regards, Timo
Jaka, thanks for the scan. I’m writing this while I’m downloading at a stunning 2.32KB/Sec! So guess I have a half hour wait….
Tiki, Thanks for finding the info, but 35MB @ 2.32KB/Sec my end…. I hear my wallet screaming – maybe I can get hold of the Siemens CD.
Charles, What’s a Cauer filter? with a simple 2nd order you get about 35dB suppression by 380KHz. The main concern is the carrier's harmonics, and interference within the AM band – which is always going to be an issue when your tuner is placed above your nice new digital amp!
Pitch254, I agree with Jaka, it would be easier to try a simpler project first, and then scale it up. I would recommend modifying the I2S on a DVD to provide the surround outputs. It’s not worth going to all these lengths to have a pure digital amplifier, only to have an extra DAC to ADC conversion loop. Your block diagram has the SRC in the “Power Stage and filter” - so no need to have a second SRC in the pre-amp. If you use a SRC, there is no need for a VCXO.
Tiki, Thanks for finding the info, but 35MB @ 2.32KB/Sec my end…. I hear my wallet screaming – maybe I can get hold of the Siemens CD.
Charles, What’s a Cauer filter? with a simple 2nd order you get about 35dB suppression by 380KHz. The main concern is the carrier's harmonics, and interference within the AM band – which is always going to be an issue when your tuner is placed above your nice new digital amp!
Pitch254, I agree with Jaka, it would be easier to try a simpler project first, and then scale it up. I would recommend modifying the I2S on a DVD to provide the surround outputs. It’s not worth going to all these lengths to have a pure digital amplifier, only to have an extra DAC to ADC conversion loop. Your block diagram has the SRC in the “Power Stage and filter” - so no need to have a second SRC in the pre-amp. If you use a SRC, there is no need for a VCXO.
triangle generator
Hi,
I have to correct my last posted simulation, the data of the triangle generator were wrong. This configuration will not be stable, because of the phase-shift of the RC-filter at the output, like the overall NFB at the digital amp. The feedback in this application should be connected directly to the output of the OPA. The connection before the integration resistor remains.
Unfortunately this configuration affects the linearity also.
The feedback resistor and the integrating resistor itself form an impedance, which is not negligible. The output voltage, growing in opposite direction than the inverting input voltage, creates a growing current through the feedback resistor - the impedance lowers. The current through the integrating resistor acts also as a discharge current for the coupling capacitor. Therefore the voltage after the coupling capacitor is influenced.
But for a linear triangle wave the integrating current through the capacitor and therefore through the integrating resistor have to be constant. That requests the input voltage at the integrating resistor to be at constant levels for high and low.
What I can imagine, is to switch the integrating resistor (input) alternating to well defined voltage levels with an appropriate low output impedance according to the value of the integrating resistor. But in this case, the capacitive coupling would not work.
A comprimise would be to increase the coupling capacitor, catching the disadvantage of long settling times.
Regards, Timo
Hi,
I have to correct my last posted simulation, the data of the triangle generator were wrong. This configuration will not be stable, because of the phase-shift of the RC-filter at the output, like the overall NFB at the digital amp. The feedback in this application should be connected directly to the output of the OPA. The connection before the integration resistor remains.
Unfortunately this configuration affects the linearity also.
The feedback resistor and the integrating resistor itself form an impedance, which is not negligible. The output voltage, growing in opposite direction than the inverting input voltage, creates a growing current through the feedback resistor - the impedance lowers. The current through the integrating resistor acts also as a discharge current for the coupling capacitor. Therefore the voltage after the coupling capacitor is influenced.
But for a linear triangle wave the integrating current through the capacitor and therefore through the integrating resistor have to be constant. That requests the input voltage at the integrating resistor to be at constant levels for high and low.
What I can imagine, is to switch the integrating resistor (input) alternating to well defined voltage levels with an appropriate low output impedance according to the value of the integrating resistor. But in this case, the capacitive coupling would not work.
A comprimise would be to increase the coupling capacitor, catching the disadvantage of long settling times.
Regards, Timo
Moderator?
Despite very nice behaviour, I´m still under moderation, and it seems hard to upload my donations to this forum. Why this is, I don´t know, but any one interested in reviewing my schematics, please mail to info@dystec.nl
Thanx, Frank
Despite very nice behaviour, I´m still under moderation, and it seems hard to upload my donations to this forum. Why this is, I don´t know, but any one interested in reviewing my schematics, please mail to info@dystec.nl
Thanx, Frank
becoming a member
Hi Frank,
my 10th posting was the first out of moderation. Your last three were inside a few hours. "It" should happen in the next time, hopefully by the next post.
A general question from my side:
I cannot find a dedicated possibility to login (Browser: Opera, starting with the actual page of the thread). Until now I start a post and switch back to the thread by clicking on the link "Anyone interested..." Does there exist an easier way?
Thank you, Timo
Hi Frank,
my 10th posting was the first out of moderation. Your last three were inside a few hours. "It" should happen in the next time, hopefully by the next post.
A general question from my side:
I cannot find a dedicated possibility to login (Browser: Opera, starting with the actual page of the thread). Until now I start a post and switch back to the thread by clicking on the link "Anyone interested..." Does there exist an easier way?
Thank you, Timo
Login
Hi Timo,
I´m using ie6 (should switch to something non MS, I know..) and I don´t have to login anymore. Must be a cookie. Then I click the link "Last page" in the thread "Anyone interested..."
The way you described seems a little complicated.
By the way, can you get hold of the class I schematics as discussed a few posts ago? Simply by reading the tekst is a little difficult to comprehend for me what is going on.
Thanx.
Cheers, Frank
Hi Timo,
I´m using ie6 (should switch to something non MS, I know..) and I don´t have to login anymore. Must be a cookie. Then I click the link "Last page" in the thread "Anyone interested..."
The way you described seems a little complicated.
By the way, can you get hold of the class I schematics as discussed a few posts ago? Simply by reading the tekst is a little difficult to comprehend for me what is going on.
Thanx.
Cheers, Frank
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.