Radial error & Decoder Reclocking...
Hi guys,
I'm happy the forums are back (with better editing fonctions) as I have no music right now. And the answer may be obvious for some of you, so I need to ask before mess in the player.
I've fitted +5V regulators around, they all work fine.
I've added a third line from the Flea trough a 100R direct to the Decoder CrIN, removed all things on the path wich were R193 and caps. Unfortunately I did not kept them in a separate box and they are mixed with others parts... and the FSM is not the right one...
When turn the player on the tray close, the focus gets OK... but the disc does not spin. Error 10, Radial error. As far as I can understand the FSM it's the Decoder which drive it, no voltage at R513 (even if it's PWM for wich I don't know if my DMM can have a look at it...).
I just hope you get that when the Decoder is out of clock...
Any idea guys?
Matthieu
(PS: if I need to go to factory set-up (and why not the older simple 100R feed between DAC & Decoder?) someone may be kind to gimme values of C521, U193, C522 and R related going to CrOut... thanks!)
Hi guys,
I'm happy the forums are back (with better editing fonctions) as I have no music right now. And the answer may be obvious for some of you, so I need to ask before mess in the player.
I've fitted +5V regulators around, they all work fine.
I've added a third line from the Flea trough a 100R direct to the Decoder CrIN, removed all things on the path wich were R193 and caps. Unfortunately I did not kept them in a separate box and they are mixed with others parts... and the FSM is not the right one...
When turn the player on the tray close, the focus gets OK... but the disc does not spin. Error 10, Radial error. As far as I can understand the FSM it's the Decoder which drive it, no voltage at R513 (even if it's PWM for wich I don't know if my DMM can have a look at it...).
I just hope you get that when the Decoder is out of clock...
Any idea guys?
Matthieu
(PS: if I need to go to factory set-up (and why not the older simple 100R feed between DAC & Decoder?) someone may be kind to gimme values of C521, U193, C522 and R related going to CrOut... thanks!)
Last edited:
Hi guys,
...i could use your advice regarding ground-layout on my 67; as i wish to optimize the ground, connecting dac and decoder to a star point (C813 main digital smoothing cap) i noticed that i may run into trouble because of the clock-coax-cables i use... .
My point is, if i run two coax-cables from my DEXA-Clock to feed the signal to dac AND decoder and connect the shields to ground, the two "grounds" of dac and decoder are connected on the clock-output via the shields, right?
Is this a problem, do i create a ground loop? Ho can i perform my "star-ground" in this case?
Reagrds,
Mickie
...i could use your advice regarding ground-layout on my 67; as i wish to optimize the ground, connecting dac and decoder to a star point (C813 main digital smoothing cap) i noticed that i may run into trouble because of the clock-coax-cables i use... .
My point is, if i run two coax-cables from my DEXA-Clock to feed the signal to dac AND decoder and connect the shields to ground, the two "grounds" of dac and decoder are connected on the clock-output via the shields, right?
Is this a problem, do i create a ground loop? Ho can i perform my "star-ground" in this case?
Reagrds,
Mickie
You should connect the coax between clock and DAC grounded at both ends.
You should connect the coax between clock and decoder grounded at only the decoder end.
If you are powering the clock from the player's main supply you would not normally connect a return ground for the supply either (this would also create a ground loop) but I think the clock you are using is different in this regard and requires it, owing to galvanic isolation of the output.
You should connect the coax between clock and decoder grounded at only the decoder end.
If you are powering the clock from the player's main supply you would not normally connect a return ground for the supply either (this would also create a ground loop) but I think the clock you are using is different in this regard and requires it, owing to galvanic isolation of the output.
Thanks Glenn!
Yes indeed the DEXA Clock has an islolated output, nothing to worry about ground loops if you use only ONE coax....
I already tried cutting the shield on the decoder's side...but things got worse (soundwise); i'll try it the other way round as you suggested.
Would you recommend terminating the cable, if so, which resistor value should i choose?
regards,
mickie
Yes indeed the DEXA Clock has an islolated output, nothing to worry about ground loops if you use only ONE coax....
I already tried cutting the shield on the decoder's side...but things got worse (soundwise); i'll try it the other way round as you suggested.
eerm, sorry i'm afraid i didn't get that, can you elaborate..?
Would you recommend terminating the cable, if so, which resistor value should i choose?
regards,
mickie
Hey guys - Nice work being done by all I think
Cheers, Lee.
A faulty player from a working one is not really nice work
Well if you try it different ways and one sounds better, trust your ears!
If you have the DAC section connected to a star ground, that's fine (if done properly) but the clock ground must be connected to close to the ground pin of the clock input.
I was just hoping your clock output wasn't grounded at the star ground, which I don't think it is, so you're OK.
Basically, do not think of gnd points as 0V. 0V is just a reference point to measure another voltage against.
If you think of gnd at the regulator as 0V then what is gnd at the DAC? If the wire connecting the two has some resistance (and it does) and it has the DAC return currents running through it, then Ohms law says that the wire develops a voltage across it (V=I x R), so there is a voltage difference between regulator gnd and DAC gnd.
The DAC is seeing the voltage between the clock input and clock ground pin (DAC ground), so that is where you want your clock input and clock ground connected. If you add in the impedance of a clock connection to star ground instead and the DAC connection to star ground, you add in the voltages created by the currents running in those grounds.
The grounds get polluted by noise currents just like the +V lines.
This is where we need Martin Clark as he's much better at that stuff and can explain it better than me too.
Anyway, if you just connect the clock input and clock ground where they tell you to, i.e., where the original crystal and capacitors were, you'll be fine because that is the best place.
Last edited:
ok, thanks, these ground "issues" are more complicated then i thought....
so if i were to optimize it, where do i start? Can you give me some spezific instructions regarding the CD67?
For example, the "digital " ground of the DAC is connected to the ground near the signal (cinch) output ground; what i wanted to do is to cut the trace and run a wire to original "star" which seems to be on C813... .
Any comments?
so if i were to optimize it, where do i start? Can you give me some spezific instructions regarding the CD67?
For example, the "digital " ground of the DAC is connected to the ground near the signal (cinch) output ground; what i wanted to do is to cut the trace and run a wire to original "star" which seems to be on C813... .
Any comments?
I would say "don't bother"!
If you don't understand 100% what you are doing you will probably make it worse, not better. That is not meant to sound like a put down because I include myself in that category. I did not implement a star ground in my CD63 (any more that is already there anyway).
The DAC digital and clock grounds go to the digital star and the DAC analogue grounds go to the analogue star at the RCA jacks, and they are both tied to the ground plane on the top of the board which joins them. They are also joined by PCB traces too, and also in the transformer.
By isolating some of these grounds you'll potentially increase ground currents in others and make matters worse. That is why I say if you can't do it 100% correctly (and I didn't feel that I could) then don't do it at all.
You could argue that you need the KI version transformer to do it properly (or at least a separate analogue TX) as only then can you separate analogue and digital grounds in the transformer to have one star in the middle of the board.
If you don't understand 100% what you are doing you will probably make it worse, not better. That is not meant to sound like a put down because I include myself in that category. I did not implement a star ground in my CD63 (any more that is already there anyway).
The DAC digital and clock grounds go to the digital star and the DAC analogue grounds go to the analogue star at the RCA jacks, and they are both tied to the ground plane on the top of the board which joins them. They are also joined by PCB traces too, and also in the transformer.
By isolating some of these grounds you'll potentially increase ground currents in others and make matters worse. That is why I say if you can't do it 100% correctly (and I didn't feel that I could) then don't do it at all.
You could argue that you need the KI version transformer to do it properly (or at least a separate analogue TX) as only then can you separate analogue and digital grounds in the transformer to have one star in the middle of the board.
Have mercy!
Please gentlemen,
I'm near to jump out of my window so please have a look at your players, try to check the values or color code of:
C522
C521
C523
R521
U193 (R193),
I wanna put things back but have no clue of wich parts were there! (Like an idiot thrown them all in the bow of no more used parts...).
It focuses, sled OK, no spin. +/-10 ok, 5Vs ok...
AAAAHhhhhhhhhhhhhhhh
MAAAAaaaaaaaaaaaatthieu
Please gentlemen,
I'm near to jump out of my window so please have a look at your players, try to check the values or color code of:
C522
C521
C523
R521
U193 (R193),
I wanna put things back but have no clue of wich parts were there! (Like an idiot thrown them all in the bow of no more used parts...).
It focuses, sled OK, no spin. +/-10 ok, 5Vs ok...
AAAAHhhhhhhhhhhhhhhh
MAAAAaaaaaaaaaaaatthieu
And unfortunately the answer is no Glenn, the FSM we all have here cover the early PCB. All that was just a 100R at that time, even the PCB is not the same.
Thanks to Ray who send me pics to help to find out.
Let's pray...
edit:
Lee, I don't have any notification either. In the profile there is an option that I've checked but with no success. Anyway the avatar picture was fixed a bit earlier, they may fix that soon too.
And many thanks for the pictures, they can confirm the ones of Ray.
Matthieu
Thanks to Ray who send me pics to help to find out.
Let's pray...
edit:
Lee, I don't have any notification either. In the profile there is an option that I've checked but with no success. Anyway the avatar picture was fixed a bit earlier, they may fix that soon too.
And many thanks for the pictures, they can confirm the ones of Ray.
Matthieu
Last edited:
Yes we can!
Yes Ray!
Lee if you had no time to take pics it's fine it's fixed with Ray's pics.
The 10p at the DAC side, the 380R, the 39p, the pink cap and a 1MO resistor to ClOut of the Decoder. A coax between ceramic caps, the sheild at the decoder side.
So my attemp to feed a third (decoder) IC with the Flea was a failure.
The last update is the numerous LM317@5V, on dedicated PSU for DAC digital, each channel of DAC analog, Decoder digital, Decoder analog, from the player TX clock 5V ref., Servo 5V ref.
Beside the fact it's no longer simple 7805s I've fitted them with a 1µF film wich "is required if regulator is located an appreciable distance from power Supply filter.". As I did not found the mark "appreciable" on my ruler...
The main gain now? Dynamic! It's fast and very invloving, let's regulate the drivers
Thanks Ray, Glenn & Lee
http://img18.imageshack.us/img18/4362/dsc00069nwv.jpg
Yes Ray!
Lee if you had no time to take pics it's fine it's fixed with Ray's pics.
The 10p at the DAC side, the 380R, the 39p, the pink cap and a 1MO resistor to ClOut of the Decoder. A coax between ceramic caps, the sheild at the decoder side.
So my attemp to feed a third (decoder) IC with the Flea was a failure.
The last update is the numerous LM317@5V, on dedicated PSU for DAC digital, each channel of DAC analog, Decoder digital, Decoder analog, from the player TX clock 5V ref., Servo 5V ref.
Beside the fact it's no longer simple 7805s I've fitted them with a 1µF film wich "is required if regulator is located an appreciable distance from power Supply filter.". As I did not found the mark "appreciable" on my ruler...
The main gain now? Dynamic! It's fast and very invloving, let's regulate the drivers
Thanks Ray, Glenn & Lee
http://img18.imageshack.us/img18/4362/dsc00069nwv.jpg
Last edited:
