QSC PL-9 Beware !

Hi,

recently, a regular customer brought in his QSC PL-9, a "9kW" beast of an a mp.

The model was first released nearly 20 years ago, as the flagship of the Po werLight range. My customer's example sat in its box until about two years ago when he purchased it and had it shipped to Sydney from the USA.

The model is no longer sold or supported by QSC and was never sold here. It s smaller brother (the PL-6 ) was sold here but is now similarly orphaned.

I agreed to have a look and found that one channel had simply stopped with no signs of serious damage - ie no blown fuses.

The amp consists of four, large modules - two PSUs and two amplifiers stack ed on top of each other in pairs and crammed into a 3RU steel box. Access i s just terrible and one needs the service manual just to find out how to ge t the modules out.

Eventually I did and still found no signs of damage, no power devices were shorted or tracks blown and it all looked fine. Next, I transferred the amp module from the dead channel onto the PSU of the working one and it ran ju st fine. So, it had to be a faulty PSU module.

The PL-9 has the weirdest circuitry I have ever seen, the SMPS is massive a nd claimed to be PFC but there is no PFC stage involved. What QSC did was t o modulate drive to the power switching stage in proportion to the AC volta ge at any moment, smoothing out the current draw but without the ability to regulate the resulting DC voltages. It works well enough, as I could see f rom the good channel. The SMPS produces four DC voltages: 50,100,150 and 20

0 volts all positive. Max current on the 200V rail is 25amps !!!

Each amplifier is a bridge mode pair with one output grounded - not so unus ual until you realise the DC supply jumps in three steps from +50V to +200V in line with the audio signal. QSC have included an elaborate scheme to co rrect the anomalies this generates and it appears to work.

My customer tried to source a new PSU module here and was told the last sel ling price was $2500 - but with none available. He might however take advan tage of QSC's cash rebate deal. Anyone who buys a new PL380 from an authori sed dealer is entitled to an US$800 rebate from QSC, if they return a PL-9 or PL-6 to the factory. No requirement that it be in working order is menti oned.

The PL380 is rated at 8kW and is of conventional class D design with no PFC in the SMPS. It is built on a single PCB and amazingly there are only *six

  • active power devices in the whole amp, compared to several dozen in the P L-9.

Anyhow, I persisted with the faulty PSU a bit longer, running it on the ben ch on its own from my iso tranny. It did not start up but I could tell an a uxiliary switching PSU was pulsing, trying to start up. Making a one turn l oop with the ground clip linked to the tip of a scope probe and dropping it over the tiny transformer I saw bursts of HF every 120mS or so. This auxil iary PSU develops a series of low DC voltages to run the daughter board on the main SMPS, +5, +/-15 etc.

If it does not run - nothing runs, so that might be the only fault. Pretty soon, I discovered 22uF SMD TANTALUM cap that read 5 ohms resistance. It wa s the input filter for a +12V regulator IC. Got it off the daughter board p ronto and fitted a regular electro in place, mounted close to the tiny tran sformer. It then ran fine and all the low voltage DC rails came up.

The main SMPS will not run unless a control voltage from the amp module is present and the manual gives a clue how to circumvent this - the LED in an opto needs to be energised. So, very gingerly I applied the probes of my ol d Fluke series 2, in diode test mode, to the relevant pins.

Blow me down, the whole shebang came to life and I had four high power DC r ails. A couple of hours more work and I had all the modules back in the cas e and wired up. Initially the dead channel still refused to start - bugger ! Then I spotted an obvious error with the AC supply to another auxiliary SMPS on the amp module. Fixed that and it came up.

Next, I performed a series of power output tests to both channels, using st eady sine waves and bursts of 4 or 8 on and 16 off with a 4 ohm dummy load. This confirmed QSC's rated output of 3.4kW per channel with 4 ohm loads. I did not try 2 ohms, where QSC rate the output as 4.5kW per channel but exp ect it would have passed.

My customer uses the PL-9 to drive six, horn loaded bass bins in his PA sys tem, each with a 600W, 15inch, 8ohm speakers.

The PL-9 and PL-6 have a reputation for poor reliability, so I hope I never sees this beast again and if it ever blows up it can stay that way, as far as I am concerned.

... Phil

Reply to
Phil Allison
Loading thread data ...

amp.

PowerLight range. My customer's example sat in its box until about two year s ago when he purchased it and had it shipped to Sydney from the USA.

Its smaller brother (the PL-6 ) was sold here but is now similarly orphaned .

h no signs of serious damage - ie no blown fuses.

cked on top of each other in pairs and crammed into a 3RU steel box. Access is just terrible and one needs the service manual just to find out how to get the modules out.

e shorted or tracks blown and it all looked fine. Next, I transferred the a mp module from the dead channel onto the PSU of the working one and it ran just fine. So, it had to be a faulty PSU module.

and claimed to be PFC but there is no PFC stage involved. What QSC did was to modulate drive to the power switching stage in proportion to the AC vol tage at any moment, smoothing out the current draw but without the ability to regulate the resulting DC voltages. It works well enough, as I could see from the good channel. The SMPS produces four DC voltages: 50,100,150 and

200 volts all positive. Max current on the 200V rail is 25amps !!!

usual until you realise the DC supply jumps in three steps from +50V to +20

0V in line with the audio signal. QSC have included an elaborate scheme to correct the anomalies this generates and it appears to work.

elling price was $2500 - but with none available. He might however take adv antage of QSC's cash rebate deal. Anyone who buys a new PL380 from an autho rised dealer is entitled to an US$800 rebate from QSC, if they return a PL-

9 or PL-6 to the factory. No requirement that it be in working order is men tioned.

FC in the SMPS. It is built on a single PCB and amazingly there are only *s ix* active power devices in the whole amp, compared to several dozen in the PL-9.

ench on its own from my iso tranny. It did not start up but I could tell an auxiliary switching PSU was pulsing, trying to start up. Making a one turn loop with the ground clip linked to the tip of a scope probe and dropping it over the tiny transformer I saw bursts of HF every 120mS or so. This aux iliary PSU develops a series of low DC voltages to run the daughter board o n the main SMPS, +5, +/-15 etc.

y soon, I discovered 22uF SMD TANTALUM cap that read 5 ohms resistance. It was the input filter for a +12V regulator IC. Got it off the daughter board pronto and fitted a regular electro in place, mounted close to the tiny tr ansformer. It then ran fine and all the low voltage DC rails came up.

s present and the manual gives a clue how to circumvent this - the LED in a n opto needs to be energised. So, very gingerly I applied the probes of my old Fluke series 2, in diode test mode, to the relevant pins.

rails. A couple of hours more work and I had all the modules back in the c ase and wired up. Initially the dead channel still refused to start - bugge r!

steady sine waves and bursts of 4 or 8 on and 16 off with a 4 ohm dummy loa d. This confirmed QSC's rated output of 3.4kW per channel with 4 ohm loads. I did not try 2 ohms, where QSC rate the output as 4.5kW per channel but e xpect it would have passed.

ystem, each with a 600W, 15inch, 8ohm speakers.

er sees this beast again and if it ever blows up it can stay that way, as f ar as I am concerned.

Hi,

Much appreciated post.

Does anyone know how I can privately get in touch with Phil?

Regards

Reply to
jo shua

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.