sounds good.
I will prepare the schematic in Orcad Capture and I will post it here for review before going into layout.
Joachim,
after some simulations, it seems that floating RIAA will work just fine.
The values on the schematic offer an extremely high precision across audio band and bass response at -3dB is 0.3Hz.
My intent is to have the circuit floating as much as possible, thus I eliminated ground reference from helpers and RIAA as well.
I would like to run it with battery only + and -.
I am currently running it like that and no ground tied to the chassis and I get no hum no external interference but sound is clearer than when tied to GND chassis and running of GND.
Please let me know your thoughts and I will include this schematic, if you guys approve it as well, on the proto layout.
after some simulations, it seems that floating RIAA will work just fine.
The values on the schematic offer an extremely high precision across audio band and bass response at -3dB is 0.3Hz.
My intent is to have the circuit floating as much as possible, thus I eliminated ground reference from helpers and RIAA as well.
I would like to run it with battery only + and -.
I am currently running it like that and no ground tied to the chassis and I get no hum no external interference but sound is clearer than when tied to GND chassis and running of GND.
Please let me know your thoughts and I will include this schematic, if you guys approve it as well, on the proto layout.
Attachments
Yeah that's what I meant! I have it work without any PSU GND.
I mean yes the GND from regulators is connected but the GND from battery not.
Sounds just wonderful.
I was thinking about this....it is a theory I haven't proven it yet.
Since I switched to the IF3602, I have started noticing that distortion .
I am running approximatively 25-30mA on 22V thus exceeding the maximum 300mW rating per side.
Maybe it is heating up too much causing distortion, don't know just a wild guess!
Last edited:
the intefet part is only on the buffer stage (see picture although values are different).
Everything else is BJT.
I am using half for the current source and half for the active device.
I am thinking to use one for the active device paralleling so each side at 15mA each, hiting then 300mW per side and having cascoded BJT current source with LEDS for bias.
Attachments
I guess 
I was really impressed by the sound that almost makes me wonder if a more complex full feedback and compensated buffer like Calvin will sound any better.
Then on the other hand, you are the expert on buffers (and not only!) and if you say that it is the best you have ever heard, then it must be really sound good.
Beside for a fully balance you need at least 4 IF3602 whjich ends up being almost $160 just for 4 parts!
I was really impressed by the sound that almost makes me wonder if a more complex full feedback and compensated buffer like Calvin will sound any better.
Then on the other hand, you are the expert on buffers (and not only!) and if you say that it is the best you have ever heard, then it must be really sound good.
Beside for a fully balance you need at least 4 IF3602 whjich ends up being almost $160 just for 4 parts!
Last edited:
I really can not tell what sounds better. The Calvin buffer is sounding better then the Paradise buffer though. Maybe it is not the lower distortion but the higher current delivering possibility. Like driving a car with a bigger engine.
Well, you started that adventure of using the most exotic parts so do not complain that it is getting expensive.
If they really bring an audible benefit i think it is money well spend.
I even think on this thread it is possible to include subjective impressions more then on the Pardise thread. This stage will be expensive to build and is more complex so it is for the chosen few only anyway.
Well, you started that adventure of using the most exotic parts so do not complain that it is getting expensive.
If they really bring an audible benefit i think it is money well spend.
I even think on this thread it is possible to include subjective impressions more then on the Pardise thread. This stage will be expensive to build and is more complex so it is for the chosen few only anyway.
ahahaha!! Yest when I wrote about $160 I realized I didn't add that statement i.e. this is a no cost-compromise design either way.
Well, the IF3602 added a definitive audible different over the simple buffer used on Paradise (only not cascoded which, again, unpopularly IMHO the cascoded version with J310 sounds worse than the version without cascode. This is just my subjective validation. In the other hand it is not like I spent different listening sessions trying with and without, so maybe on long term listening I would end up preferring the cascoded version over the normal one).
If no cost objective is a concerned here, which it shouldn't, then I would use 4 IF3602 per channel each one biased at 15mA where current source is made with half of the IF to match better thermal drift and output offset; then move the servo before the buffer.
I will give it a try to it tonight see if:
1) gets rid of the distortion
2) sounds any better than only one
Maybe you can add one of these EEVblog #460 - Mailbag - Plasma Speaker - YouTube it would be a good hi-(Voltage)-end component.
This is probably the version I am going to build if it's ok with developers here (Joachim, Franz, MiiBi).
Since it is not cost objective design, the IF3602 proved to be that much better over the "standard" paradise buffer.
This doesn't mean it will be on the final design as once proto is in a comparison with other buffer will be run.
If so I will try to enter this on Caputure hopefully tonight (however need to try the two paralled IF3602 at lower current to verify that sound is mantained or bettered and if distortion is gone, so time might be a bit tight).
Once schematic on Capure is completed I will post here so we can all review it before going to layout.
Here is what I have on the simulator so far. This is really an exciting design.
I will also try in the next days to see if phono can operate without ground from regulators thus with a complete floating audio ground.
Since it is not cost objective design, the IF3602 proved to be that much better over the "standard" paradise buffer.
This doesn't mean it will be on the final design as once proto is in a comparison with other buffer will be run.
If so I will try to enter this on Caputure hopefully tonight (however need to try the two paralled IF3602 at lower current to verify that sound is mantained or bettered and if distortion is gone, so time might be a bit tight).
Once schematic on Capure is completed I will post here so we can all review it before going to layout.
Here is what I have on the simulator so far. This is really an exciting design.
I will also try in the next days to see if phono can operate without ground from regulators thus with a complete floating audio ground.
Attachments
You need to connect the servo to the output of the buffer, otherwise vgs (when connected to the input (as shown
)))) of the fet's is used to create an imbalance in the mirrors. And if you insist, the experiment (that fails / will fail) will be to switch it to the buffer input (and back (very fast)).Edit: Wrong I was. I should have said that the vgs of the fet's would be translated in a output offset if the servo-input is connected to the buffer input.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.