Sony TA3200f Issue

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Administrator
Joined 2007
Paid Member
OK, that's no problem. The amp will perform correctly even if the current were different so don't worry to much over that over.

If you have the old diodes then you could connect a series resistor to one and connect it across a 9 volt battery to test and measure the volt drop. The resistor would need to be in the region of 4.7K to 8.2K (ish) to give a nominal 1 ma or so. It would be interesting to know for certain that they really were as drawn. Its only the marked voltage on the diagram that made me wonder, thats all.

Great result though, well done.

(that's me done for tonight now :))
 
congrats must be immensely satisfying to have this sorted :)

Thanks kasey197 couldn't have done it without you guys. Great tip on the freezing, really did the trick. Mooly was inspirational and nice to know there is a like minded community out there that will stay with you:)

Have to say this amp is worthy of the attention it is getting. The soundstage is both wide and deep, very deep yet controlled bass response, highs are very palatable with a sumptuous mid range. It has power to spare as I am running a vintage pair of Braun L810's at 4ohms. I rebuilt the X overs last year and they are a great match for this amp. The main filter caps are Vishay Sprague's which was definitely worth the price of admission.

Back to restoration...
I still have 2 1N4148 diodes left. Just wondering, can I apply the same method to the PSB D305. It also has one of those nasty diodes. Put 2 in series. I will measure the voltage before I yank them Mooly:eek:

Since I will be replacing a number of components next week I was wondering about this big ugly metal resistors on the Power Amp board R121 through to R235. They are .22 5W Metals. I thought a good substitute would be Mills Wirewound.
Mills MRA-5 Resistors Thoughts on this.

I thought sonically this may be a good improvement.
 
Administrator
Joined 2007
Paid Member
That's quite an unusual bit of circuitry around D305. It looks to be configured to help supress switch on/off noises as the rails rise and collapse. Yes, you could replace it in the same way but :) 1N4148 types may not be the most suitable due to their very small current handling capacity. I'm just wondering what the ultimate surge currents would be worst case. I'd probably play safe and fit something like 1N4004 or 1N4007 Also, measuring the voltage across these wont work because they are reverse biased in normal operation. In other words they only conduct as the rails collapse.

R235, 0.33 ohm is one of the amps secret weapons and helps with the ultimate stability of the whole power amp into capacitive or reactive loads. I would say you would gain nothing tbh in replacing them.
 
Member
Joined 2010
Paid Member
O/T - I'm also curious about the design of the input stage with Q102 and the bias current balance in the LTP.
The term "Paraphase Amplifier" on the schematic may be just marketing terminology but obviously someone
thought it was worth more complexity than the standard LTP.

There's little information on the net so I wonder what the plan, apart from just trying to be different, was? :scratch1:
 
Member
Joined 2010
Paid Member
........Since I will be replacing a number of components next week I was wondering about this big ugly metal resistors on the Power Amp board R121 through to R235. They are .22 5W Metals. I thought a good substitute would be Mills Wirewound.....
Mills MRA-5 Resistors Thoughts on this...
Both types are "low inductance" wire wound resistors. First, you should consider the fit and lead spacing which, with Mills' solenoid coil format, seems better left as is. The only benefit would be the cute olde worlde appearance but once the cover is on, there is nothing to show or hear for the expense.
 
That's quite an unusual bit of circuitry around D305. It looks to be configured to help supress switch on/off noises as the rails rise and collapse. Yes, you could replace it in the same way but :) 1N4148 types may not be the most suitable due to their very small current handling capacity. I'm just wondering what the ultimate surge currents would be worst case. I'd probably play safe and fit something like 1N4004 or 1N4007 Also, measuring the voltage across these wont work because they are reverse biased in normal operation. In other words they only conduct as the rails collapse.

R235, 0.33 ohm is one of the amps secret weapons and helps with the ultimate stability of the whole power amp into capacitive or reactive loads. I would say you would gain nothing tbh in replacing them.

So we will leave the big metals alone. Thanks for your input as well Ian.
 
Administrator
Joined 2007
Paid Member
So we will leave the big metals alone. Thanks for your input as well Ian.

Best bet I think for those :)

(for the sake of completeness (but don't feel you have to do this), it would be informative to know what those replaced diodes really did develop in the way of a volt drop. This shows the discrepancy in the marked voltages and the 1.4 volts that two diodes generate)
 

Attachments

  • Capture.JPG
    Capture.JPG
    71.6 KB · Views: 74
Best bet I think for those :)

(for the sake of completeness (but don't feel you have to do this), it would be informative to know what those replaced diodes really did develop in the way of a volt drop. This shows the discrepancy in the marked voltages and the 1.4 volts that two diodes generate)

I would then
1. take a 4.7ohm resistor, attach it to one end of the diode.
2. attach remaining resistor lead to a 9V battery
3. attache VM to diode and then to the other terminal of the 9V battery.
4. Take reading which should be 9V - 1.4V for a total of 7.6V

Have I got that right.
 
Like this

Still have to get the 9V battery Mooly.

Run into a bit of trouble. I replaced the tants with 2 small lytics and all is good.
I have now moved into replacing the carbon resistors. After about half of them installed I thought I should test to see if thing are good. I get the old dim bulb tester to be on the safe side and presto a short.

I am using Takman metal film 1/2W resistors. These should be a good match. I have gone over all my solder joints and inspected for pulled wires, checked values and all seems to be good. What do I need to do to fix this?
 
Administrator
Joined 2007
Paid Member
That could be anything.

What resistors have you replaced ? Have you been working on one or both channels ?

I would recheck all your work and check the new resistors are what you think they are value wise and that they are in the correct place in the circuit.
 
OK! Nothing like slowing down and taking a pause. I checked it all and yes R110 and R210
call for a 47k and I had a 47R instead. I put the old ones back in as I have no replacements for them on hand. No wonder Mooly, that's a huge difference. Fired it up and the bulb went dim. I'll get the hang of this sooner or later. 110 and 210 are both out of spec though.
Just hope I haven't messed other things up.
 
Administrator
Joined 2007
Paid Member
Its a good job you used the bulb tester.

R110/210 are part of the constant current source and "pull" current through the diodes you have just replaced. I would be inclined to replace those again after that because you are bound to have exceeded the maximum specified forward current. Nothing else will have suffered fortunately... which is good :)

When you say R110 and 210 are out of spec, do you mean the new or the old ? I'd be very surprised tbh if the originals had gone high.

Those 47 ohms that took all the current should be checked too.
 
Its a good job you used the bulb tester.

R110/210 are part of the constant current source and "pull" current through the diodes you have just replaced. I would be inclined to replace those again after that because you are bound to have exceeded the maximum specified forward current. Nothing else will have suffered fortunately... which is good :)

When you say R110 and 210 are out of spec, do you mean the new or the old ? I'd be very surprised tbh if the originals had gone high.

Those 47 ohms that took all the current should be checked too.

I replace all the dubious resistors Mooly.

On another note, I decided to replace the old RCA inputs. There are 2 sets which is nice. While I was at it I replaced the signal wire with some Mogami shielded cable. This goes from the inputs to the source switch. This upgrade alone is probably the second most notable upgrade to the sonics I have done so far. Big difference. This bring me to this question, which maybe for another post, but here goes anyway. The manual says you can go direct with your source into the 2 inputs. There are L and R channel level controls. They don't use the term Passive Preamplifier, but in effect that is what this is if I am not mistaken. I have tried this with both the CDP and phono and am rather impresses with the sound. I have even more gain from the CDP which in effect is very loud indeed (2v out) and considerably more from the phono, which is nice since I was about half volume on my Krell KSP7B pre to get a decent level for listening and much less going direct.

So the question is, is there any way of upgrading the level pots to something higher end. alps ect. Should it be an attenuator or just a better quality pot, which I am having a hard time finding. 150ohm pot.
 
Last edited:
Administrator
Joined 2007
Paid Member
It's a 150K :)

That's interesting because today we would use much lower values due to preamps and modern source components having much better drive ability.

You can use anything value wise here. I would say 10 to 47K (50K) would be ideal. A real downside of high values is that the pot and circuit capacitance it sees on its wiper form a variable low pass filter (an R/C filter) that can affect the frequency response as the pot is rotated.

If you don't need two inputs then I would bypass the switching and route one of the inputs direct to the volume controls.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.