Roland KC-880 Keyboard amp...no output, no service doc's

[nevetslab]

Q4 on the schematic you posted is RN1442, not 1422.

OPPPS!! I thought for a moment it was just a typo, but after I checked my Mouser invoice, I see I did order the wrong part. Further clue was the letter code on the part I installed. I just ordered the correct part from Digikey, so I'll have to wait until next week to see if that solves the problem. Odd that I still had the correct MODE potential, that being in Operate mode. Hopefully changing to the correct part solves this. Sigh.........

[nevetslab]

Have you got both mute and standby voltages when requested?
Have you got oscillator out on pins 16 and 21? VSS and VDD ok? Check you have audio on pins 8 & 4.
The TDA is an extremely stable chip that shuts down when there is a problem; over volts, under volts, over current etc.

The Speaker MUTE button in the AUX section doesn't change anything still, so it may be due to my selecting the RN1422 part instead of the RN1442A part. I'll have to set this aside until the correct part arrives.

[nevetslab]

Top of the YEAR to ya all!

Now that I'm back in the shop, having driven here yesterday, only to find the facility still locked up tight, I learned later that day after emails that nobody bothered to inform me CenterStaging wasn't opening until Jan 4! So now, still waiting for the arrival of a small package from Digi-Key, my test bench is occupied by the contents of the Roland KC-880:





That surface mount part RN1442A Q4 is underneath the Power Supply/Power Amp PCB assy. As you can see the lead spacing on the TDA8920B Class D Power Amp IC's are on very tight lead spacings, with the second row of pins underneath the package, so really only accessible on the bottom side of the PCB for probing. I'll have to unmount the PCB assy, lay down an insulator sheet so I can turn it upside down (after I first replace Q4 with the correct part). Not so service-friendly.

KC-880_sm_PA-PS_Pg 1.pdf

[nevetslab]

Well, the Digi-Key order arrived, now have the correct part installed for Q4 on the bottom side of the Power Amp/Power Supply PCB. Re-connected everything, powered up, and I still have the same power supply consumption readings of 0.3A/28.5W @ 120V, but NO output. Checked the MODE voltage, and watched it from Start-up (@ 0V, then ramps up to +4.4VDC. The PWM Oscillator is a separate circuit, rather then using the internal Oscillators so the three Class D amp IC's are in sync. 392kHz with 5V P-P, and offset from Signal Gnd as per the ap notes in the data sheet for the TDA892B IC. Checked to see that the SPEAKER MUTE button actually mutes the output from the Control panel. It does....mutes all four signals (L & R LF, L & R HF), so that circuit is working. All of the 4A power supply fuses are intact on the PA/PS PCB. I loaded the Left Ch this time with the other woofer/tweeter, so I have +5.1VDC on both Outputs of the Left Ch woofer (no differential voltage between outputs), +0.2VDC on the Left Tweeter, +0.7VDC on the Right Tweeter, and _11VDC on both Right Ch woofer outputs (no differential voltage between outputs). Not sure why there would be substantial offset voltages on the two woofer IC's, relative to GND.

I haven't yet removed the PA/PS board, turned it upside down to see what I read on the Class D IC Pins. That's next. I've attached the individual service manual schematics & PCB diagrams, less the rear panel input jack panels/circuits. All of the inputs and jacks work, so they don't appear to be involved in whatever is inhibiting this stupid thing from working.

KC-880_SM_Block Diagram.pdf
KC-880_SM_Control Panel diagram-1.pdf
KC-880_SM_Control Panel diagram-2.pdf
KC-880_SM_Main PCB-1.pdf
KC-880_SM_Main PCB-2.pdf
KC-880_SM_Main PCB-3.pdf
KC-880_SM_PA-PS_PCB-1.pdf
KC-880_SM_PA-PS_PCB-2.pdf
KC-880_sm_PA-PS_Pg 1.pdf
KC-880_sm_PA-PS_Pg2.pdf

[nevetslab]

Yesterday I finally got the Power Supply/Power Amp PCB removed from it's chassis plate and turned over, while remaining connected to the control panel and Power Xfmr. I powered up, and recorded the voltages on the three TBA8920B Class D Power Amp IC's. While I found the VDDP1, VSSP1, VDDP2 & VSSP2 voltages all measuring +/- 27.2VDC, the readings thru the 10 ohm resistors to the Vdda1, Vssa1, Vdda2 & Vssa2 Analog voltages for the front end were all over the map, AND.....measuring those 10 ohm resistors R5, R6, R7, R8, R35 & R36 were orders of magnitude greater than 10ohms. AS a result, suspecting those 10 ohm resistors are open circuit or something not even related to 10 ohms (SMT Case Size 0603....1.6mm x 0.8mm, 1/8W), each of the Power Amp IC's are in Unbalance Protection Mode, and shut down. I did NOT have any loads connected to the Output wires (that feed the woofers and tweeters).



This is, of course, Lead Free Solder, and the soldering of the IC Pins of there three Power Amp IC's doesn't inspire confidence either. Lots of solder blobs all over the PCB top and bottom. I suppose I could try replacing those 10 ohm resistors from the +/- 27V Supplies that feed the +/- Analog supply terminals Vdda1, Vssa1, Vdda2, Vssa2 and see what happens. I had readings all over the map for Boot1, Boot2, PROT, & STAB1 on the Power Amp IC's. No idea what caused this problem, so just replacing those 10 ohm resistors, my suspicion is they will just fail again.

Ain't this fun??!!

[nevetslab]

I finally received the flea-sized 10 ohm 1/16W 0603 size Surface Mount resistors that feed the +/- 27VDC to the Analog Side of the TDA8920B Power Amp IC's. PITA working that small, but got all six resistors installed (R5, R6, R7, R8, R35, R36.). I connected the power xfmr and control panel wiring to the amp module, set up much like the photos above with the PA/PS PCB upside down. Connected the speakers & tweeters, just in case I got lucky.

I briefly saw around 80WAC Mains and +/- 27VDC passing thru the new resistors. But, as I suspected....a tiny trail of smoke began coming from underneath, and the power reading dropped as reached to shut it off. Checked for the presence of those new resistors. All but two were open circuit. Why am I not surprised?

I had found a seller here in the US that has a few of these NOW-Obsolete Power Amp IC's and bought them for just under $37. So, this project drags on, and the road case with the amp pulled apart sitting atop the case just barely fits into the shop. Can't get in until I roll it out in the mornings.

Sigh.............

[nevetslab]

The replacement TDA8920B Power Amp IC's arrived yesterday, so after setting the Ampeg SVT-CL aside, I pulled the power amp/power supply board back off of it's chassis mtg plate, then loaded the Desoldering Iron with a .031" ID tip and first chopped off the leads of the power amp IC's to get those bodies out of the way. Then, went thru the tedious desoldering procedure of unsoldering each pin, removing the stubs from the the top side of the board, while getting as much of the solder sucked out of the pads on the first pass. Double-sided PCB, of course, so it still took several passes to get all of the solder out of the holes while not damaging any of the pads, and finally had clear thru-holes for the replacement Power Amp IC's. I got those installed & mounted, then soldered them in this morning.

With the Power Amp PCB still unmounted, I re-connected the pwr xfmr, speakers/tweeters and control panel together, along with a DMM and the enlarged pin-out of the power amp IC's, and powered it up. The power consumption is still sitting around 30W as it had been before, and....nothing coming out of the amp when fed signal. Why am I not surprised.

I went thru and recorded all of the voltages on the three Power Amp IC's, and at least this time, no DC offset on the outputs, all of the voltages appeared nominal, but, I did find -18VDC on each of the Protect Pins on the chip. I don't find any information on the data sheet for what that voltage would be. All of the supply readings on the chips are equal/opposite, Boot DC readings nominal as called out in there data sheet., so this time, while not working to pass audio nor pulling current that would reflect the nominal power consumption readings (if that nominal 90W is at idle?), this is still dead in the water. Mode pin still sitting at 4.4V, and follows the start-up sequence. Oscillator reads correct (ext osc circuit, sitting at 370kHz, within the operating range of the chip).

That's enough horsing around with this unit. Not built for service, equipped with now-obsolete Power Amp IC's, and while everything thru the send circuits on the control panel work, those signals are NOT getting thru this Power Amp / Power Supply PCB assy to work. I admit defeat.