Panasonic class D amps

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
a "clicking" or chirping switcher means that it is in shutdown mode, triggered when the switcher's feedback rises above a certain setpoint, telling the switcher chip that too much current is being drawn. this could be happening because of heat, or some circuit condition that looks like a short (at least as far as the feedback to the switcher chip is concerned). this receiver has a lot of separate interdependent power supplies, and to run this beast off of batteries could create a lot of headaches. as i mentioned the simple solution is to wire up 10 gel cells in series to a power outlet box and plug the receiver in to that.
i'm not yet absolutely sure but one or more of the transistor switches shown with input signals from the cpu, may need to be switched on or off in proper sequence by the cpu for the receiver to power up properly. many times power supplies in these receivers need to be timed in their startup sequence to avoid damaging things like PWM amp chips and regulator chips.
 
I forgot about the last bit of the PS schematic - this part seems to be the DC Detect circuit?

Oops, Unclejed, crossed posts - OK.

But are 10 12V cels going to develop 120V DC instead of AC? Also I wanted to get rid of the switcher noise that I'm sure is on the PS!
 

Attachments

  • xr57 ps part 3.pdf
    67.3 KB · Views: 182
i would guess that the switcher noise is either a) buried in the noise of the PWM amp, or b) if it does appear at the outputs in any form, it's filtered out by the speker filters. also, without the 120hz ripple on the primary side of the switcher, the modulation effect that makes switcher noise more audible (the intermodulation of the switcher frequency and 120 hz harmonics), there is far less switcher noise.
 
Ok,

I finally connected the D-Clock into the XR-55 (same connection for the most part in the XR57). I used 15vdc from the power supply board, think it was W302 jumper towards the front of board/unit.

Results?

The sound has really skyrocketed! My wife in the other room even said something. So much more open, much clearer, more detail and nuances. I know my sound perfect and now its really escalated to a much higher level! Much more finesse and delicacy to all instruments, everything sounds so much finer.
Why didnt I do this sooner?

That crystek is not going to really give you that much of any noticeable improvement in my opinion. The price of the clock I used cost more than buying another Xr55!
 
Good work Crazy_D,

This is the sort of improvement I found with running off batteries & when I tried the crystek clock, I found no change

Did you try batteries on the output stage - if not give it a try ( a much cheaper experiment than your clock) and let us know?

I may give another clock a try!
 
Hi,

Thanks for the response. Yes it was a great improvment. Everything here really became ultra clear. The clock isnt even burned in yet either.

Are you going in digital via coaxial or toslink? I modded the DVD input coaxial circuit to the akm chip, by replacing resistors (75ohm and 47 ohms with caddocks) and replacing the 0.1 coupling caps with blackgate 0.1uf/50v NX caps, and did a hard wire silver wire connection point to point to avoid the ribbon card cable.

Whats up with that inverter chip at the coaxial inputs? Whats up with the input chokes as well? I wonder if those can be either bypassed, or if not, replaced with something better?

No I havent tried the output stage on batterys, i take it I solder sense + to Vdd and sense - to amp ground on the output stage board side and then wire those over into a high performance battery with some large value cap(s)? I have a optimal yellow top just sitting here and a boatload of capacitors.

Please clarify the output speaker wiring and binding post mod. Did you just remove the whole speaker binding post circuit board, installed your new posts and run new wires, soldered to the board, from the outputs from the amp board? (bypassing the relays?)
What about output caps?
Are you using any to ground at the output or does it not need them? I beleive their are some on the stock board.
If you are using them, what value?

Lets try to rekindle this thread! ;)
 
I'm using digital co-axial in. There is a digital input circuit here from Jocko Homo which is reckoned to be excellent - he knows a thing or two about digital circuits (it's all got to do with avoiding signal reflections). I haven't used it but might do so (can't find the link at the moment). My thoughts were that all the XR spdif input circuitry could be bypassed with Jocko's circuit.

I have disconnected the PS wiring between the PS board & output board Rewired the sense wires as you said & then connected the 2 SLAs for 24V through 10,00uF cap to the + & gnd on the output stage board. Really cleans up the sound & the vol is still very useable (it just has a plateau in the middle vol range) I would be very interested to hear what further improvements this brings to your system?

Edit: This is the biggest weakness with non feedback equibit amps - they have very low PSRR so PS is critical!

I did as you said for the speaker wiring - actaully left the speaker posts in place but disconnected their wiring to the output board. I then solder wires to the speaker posts directly from the output board This bypasses all relays & bad wiring, etc. I used no caps on the output.

I agree that there is more to be wrung out of this amp so let's do some more investigations!
 
most likely it is. most receivers today are mostly digital, with the only analog signal chains from the inputs to the DSP chip, then from the DAC outputs from the DSP through some low pass filters and into the power amp stage. even with class D, most receivers are constructed this way. the way you can figure this out is to see if the receiver has an "analog bypass" or DIRECT function, if so it has a "semi analog" signal chain. if a receiver doesn't do this, it's entirely possible that the DSP chip could be driving a class D amp with direct PWM from the DSP chip. this would be a CODEC function meant to eliminate any analog signal chain (except for the inputs and the speaker outputs).

with a 0.7% distortion figure, i would personally pass on this one. class D is nice and efficient, but not a whole lot of commercial D amps are in the category of "Hi-Fi". manufacturers cut corners by not using any NFB methods.
 
Hi,

Maybe someone involved with this thread could help. Actually, I ask the similar question on separate topic but nobody replied. I am using xr55 in stereo mode only, fed through analog cd input. Receiver currently works with 4 amps – stock dual amping (2 amps working together per channel) and I would to like to add one more amp per side, so 3 amps to work together, i.e. total of 6 amps will work in stereo ( 2 x 3 amps per channel).

Looking to the xr55 manual, page 8, bi-wiring connection, I wonder, is it possible to connect plus speaker terminal for HF and plus for LF, and the same with negative terminals. Same connections, to be made separately for left and right speaker.

Anyone have idea?
 
you mean you want them bridged? which if they're already bridged outputs to begin with operating from common power supplies, is going to "let out the magic smoke". if you mean you want to parallel their outputs, then you need an isolation filter between the paralleled outputs to keep the HF portion of the output from "contention" and cross conduction. i don't have the schematic in front of me, but IIRC, each amp channel runs off an independent PWM clock (samsungs run off a master PWM clock, Yamahas don't, and i don't remember if Panasonics do or not). paralelling could also run into problems with the production of beat frequencies from the nonlinear mixing of the HF components if each amp has it's own clock. bridging would have to be done with amp modules running off a common clock and isolated output stage power supplies
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.