GK 550 Fusion full PS rail pulse at turn-on, then goes to 0V

[nevetslab]

Finally back to Back-Line gear maintenance. Going thru the GK bass heads from our NY Inventory, and found a GK 550 Fusion which, when first powering it up, output connected 4-wire to a GK 410 Bass Cabinet, I heard low-level hum while the Red light around the power switch is lit, then changing to Blue light, I heard a substantial thump thru the speaker, amp in Standby mode (Blue light flashing). I unplugged the speaker cable from the amp, and plugged my Fluke 8060A DMM into the 1/4" speaker jack. With the amp off, I was reading -0.67VDC. Interesting. Switching the amp on, that changed to -6.7VDC while the Red lamp is on, then while switching to Blue light, I briefly saw the output read -74VDC before immediately going down to nominal 0V (around 12mV offset with the amp on.) I repeated this several times. Each time, during the start-up powering, reading -6.7VDC, a quick pulse to -74VDC, then down to 0V. I fetched my adapter cable to allow reading both Woofer and Tweeter, and checked to see if there was anything similar occurring on the Tweeter amp output. 0V during power-up, 0V during powered on, but a brief -3 to -4VDC pulse during power-off, which immediately dropped to 0V. No load (10M) when just connected to the DMM.

I can't say I've seen this behavior before on any GK1001RB's or GK550 Fusions. The GK 550 Fusion uses the power amp PCB assembly from the GK 1001RB. I still haven't called GK to have them send me the service documents for the 550 Fusion amp. I've attached the power amp PCB Schematics for the GK1001RB for the time being.

1001RB-II_POWER_AMP_PN_206-0260.pdf

[nevetslab]

Yesterday afternoon, I removed the power amp PCB assembly, and went thru it this morning in hopes to find the cause of this short DC output pulse. I was mistaken...the short (~50mSec) pulse goes positive to around +84VDC, then drops rapidly back to 0VDC (takes maybe 200mS to fall to 0V).

I checked all of the output stage, driver stage, front end, etc for any failed xstrs, as well as all the diodes on that PCB. Checked all the resistors, electrolytic caps (400mV 5kHz Sq wave pulse used for ESR tests), not finding anything wrong. The Woofer Amp does pass signal just fine, though the tweeter amp didn't pass signal. Disconnected the Woofer and Tweeter signal cables from the preamp without any difference. Tube plate supply disconnected also made no difference. So, put it back together, finding the same symptoms. So, wrote up a short description of the problem and will have our staff ship it back to GK....it is their equipment anyhow.

As luck would have it, I had set up my Tek R5031 Storage Scope to capture that turn-on offset pulse and then take a photo of it, but, before I was able to complete waveform capture, the scope up and quit on me. Power supply failure. So, another maintenance task to add to the list. That was the only storage scope I had here (for the moment). I guess I could bring over the Analogic 6000A Digital Scope over and see if that's working ok.....parked under my test bench at the main shop down the street.

[Jazz P Bass]

If the positive pulse shot up above the low voltage rail (+43.5Vdc) wouldn't that indicate that the high rail was being turned on? (briefly)

[nevetslab]

If the positive pulse shot up above the low voltage rail (+43.5Vdc) wouldn't that indicate that the high rail was being turned on? (briefly)

So it would appear, looking at the schematic. Until signal is driven above the potential where the lower set of outputs are discretely fed thru the MUR410 diodes of the lower supply rails, the upper devices in series aren't turned on. Today, I've been moving GK 1001RB amps from the New York inventory shelves thru the test bench, first checking the DC levels of both the woofer and tweeter amps prior to connecting to my GK 410 test speaker, to see if there are any trends to follow. All of our GK 1001RB amps don't have any appreciable DC offset voltages occurring turn-on, at least that which can be seen by the 2.5 reading/sec Fluke 8060A DMM. Once the amp has been powered up, I'm finding the woofer amp, after being turned off, there's a nominal -0.23VDC potential on it's output, which slowly drains down to 0V. The Tweeter amp, during turn-off, remains near 0V, then quickly flies down to a bit over -5VDC very briefly, and rapidly returns to 0VDC. So, there is a trend on that negative DC shift on the woofer amp that's common to that faulty GK Fusion 550 amp assembly. I haven't seen or heard any of these amp exhibit that high output pulse though.

As this defect on the Fusion 550's woofer amp is a quick transient, it doesn't surprise me that nominal low current tests with a DMM in Semiconductor junction tests won't show a device that has some character that's allowing that short conduction and turning on one or more devices on the positive side upper stage.

I've never tried to set up a test fixture or a set of cables to operate either the 1001RB or 2001RB power amp PCB assemblies outside of the amps' chassis to try and probe the output stage, driver and front end stages. Everything in these amps is tightly packed together, and all of the output xstrs are neatly mounted underneath the heat sink. Nice packing job as far as that goes, but not service-friendly when you find something like this occurring.

Having found there are presently five GK 1001RB amps in the NY inventory, with one of those having a yellow gaffer tape note on it's cover, indicating that particular amp was sent out to Blackstone as a loner 'a long time ago', and has been parked in their inventory for some time. I just asked our Guitar Dept manager if he'd mind if I swap the power amp PCB assembly between this 1001RB and the defective Fusion 550, since the same power amp PCB is used. He too saw the logic to that, and, whenever that amp is returned to GK, the shipping weight would be much less than the Fusion 550. Conveniently restores the Fusion 550 to the working inventory, so I got the ok to proceed.