I've just acquired a faulty Crown 800CSL. The guy I bought it off said that it worked fine until somebody decided to give it a drink...
On initial power up test, I found a 60V (approx) DC offset on Ch1 and about 32V DC offset on Ch2.
When I took the amp apart, there wasn't much evidence of liquid but there was evidence of somebody disassembling things at some stage, eg. wire colour markings on heatsinks. There were some burned resistors on the output PCBs too and Ch1 seems to have suffered more than Ch2.
On this basis, I completely disconnected Ch1 and set to work on trying to get Ch2 back in a working condition. I have completely rebuilt the output transistor banks on Ch2. One device only had 1 leg (the 2nd leg was completely missing - suspect it got broken off when the previous repairer had a go at it) so I have used a device from Ch1 instead. All drive and output devices pass a basic transistor test with a multimeter, so I'm happy that they're ok. I have replaced R01 and R10 (both 470 ohm) which were both burned out.
The problem is, I still have about 30V DC offset on that channel after the power-on delay kicks out, when the main module is plugged into the output module. If I just run the output module up, there is no DC offset. If I use Ch1's input stages from the main module, the DC offset goes. However, there is a buzz on the output and it is impossible to adjust the biases to the correct voltage (0.3V). The high side bias won't change from 0V and the low side bias will only go down to about 0.45V.
I did also notice that R30 (programs S100) has become quite hot, although still measures 102 ohms. If S100 (LM224/LM234 current source) has also been fried, could that cause the problem I'm having? Is it possible/safe to disconnect the ODEP circuitry completely, to rule that out of the equation whilst I have it on the test bench, or should I be looking somewhere else?
One thing that did concern me is that the DC offset protection is obviously not kicking in. The theory in the service manual talks about "control/protection transistors" which are "tied into the voltage translator inputs", which "act as switches to totally shunt audio to ground during the turn-on delay, or during a DC/LF or fault protective action." The turn-on delay works, but the DC offset protection obviously isn't.
Just thought of something: I also have fully working 460CSL and 1400CSL units. Could I borrow the main module out of either of these (the 460CSL is fitted with 43060-7 and the 800CSL has 43059-7) so that I can narrow the fault down to the output module or main module? I remember Gary at Fuzion in the UK writing that the main modules of this design of amp (Micros, Macros, CSLs, Geodynes, etc.) were largely compatible with each other - for testing purposes if nothing else.
Edited by Wrighty, 10 March 2005 - 10:02 AM.