Krell KSA 50 PCB

Comp1
 

Attachments

  • pcb2.jpg
    pcb2.jpg
    73.4 KB · Views: 960
Hi,
have a look at the trace joining E$37 to E$9, particularly through the HF cap used to bypass the two electrolytics.

I believe that route should be short and straight.

Apply a similar logic to other routes and see if any can be shortened and/or straightened.

The KSA50 has no DC blocking on it's input. You have a lot of spare PCB space at the input side. Do you want to consider adding pads and an alternative input pin for a large (wide pin pitch) polypropylene blocking cap for any future source that requires it?

Do you consider Temperature equality between the LTP pairs to be important for good DC characteristics? If so then the pairs of input transistors should be facing and touching each other.

The power supply rails are spaced, as many others do at the opposite extremes of the PCB. Would you consider placing them adjacent to each other?

Have a look at the trace from E$20? (the NFB bypass cap) to the input ground. It picks up a series of traces along it's route, the complete opposite of star grounding. Is this a factor you want to consider?
 
Here’s another version, an offshoot of my Leach layout. The VI limiter dual slope arrangement of Mikeks’s is again fitted, but not shown on the cct diagram. This requires at least an extra 26 components for a two pair output, so it’s not surprising that most pcb layouts forego it!
 

Attachments

  • krell50.zip
    83.3 KB · Views: 305
There are still 50 to 60 KSA-50 boards left (Pinkmouse version). Boards are $19.00 each plus shipping. Domestic shipping is 4.95 and overseas shippping is 9.95. Italy is $12.95 because it requires a special envelope I have to buy. I take PayPal or you can send a check or money order. E-mail me at cinerama84106@removeyahoo.com. Be sure to remove the remove in the address:) .

Mark
 
I am now trying to resolve a problem. Before connecting the output devices, I wanted to smoke test the boards. Using a bench PSU, I connected the boards to about +31.6V/-31.6V. Looking at the measurements posted on the thread, I believe something is wrong. These are the voltages I found relative to ground.

Q E B C
Q-101 -1.17 -0.561 30.86
Q-102 0.075 -0.561 -28.15
Q-103 -1.03 -0.364 28.3
Q-104 0.248 -0.364 -30.74
Q-105 28.94 28.3 0
Q-106 -31.62 -30.73 0
Q-107 28.94 -2.3 16.93
Q-108 -31.29 -2.3 -31.61
Q-109 -2.3 31.63 -1.84
Q-111 -2.3 -31.63 -1.85

These are the voltages taken across the resistors, as presented in the Wiki page.

R-103 -31.05
R-107 31.05
R-108 4.42
R-109 5.23
R-112 0.77
R-113 3.4
R-116 3.3
R-117 0.89
R-120 2.69
R-121 0.33
R-122 0.92
R-123 0

Any help will be greatly appreciated.
 
Loek,

Thank you for your suggestion. I have double checked all the components and the values are correct for their positions. I’ll try replacing the transistors tonight when I get home. :Ouch:
If it isn’t too much trouble, could you walk me trough the steps you took to reach your conclusions. I am hoping to learn more about troubleshooting this amp.
 
What i posted is not so difficult to follow Grimberg.
If you take the schematic in front of you it is clear that some transistorlegs are direct connected to the next one and must have the the same voltage.
I use the Jan schematic, (Delta-audio) and there E-Q105 is connected with B-Q107. Not the same in your table....
There is also a Q110 missing in the table...
Maybe you made a mistake in writing and reading voltages.
Loek