Phase Linear 700 Series Two Help Please

[Mbose]

Hi, I joined a couple years ago, with the expectation of picking up some Phase Linear gear. Took a lot longer than expected and appreciate any help you guys can offer with a 700 Series Two.

When I got it, the 700’s main fuse was intact, but the rail fuses were blown and wrapped with a double strand of 32ga copper wire. I try to make up for my lack of knowledge by being cautious, and decided to poke around before trying variac/dim bulb startup. I got the service manual and would like to share findings and photos here, along with some questions.

• The photos show new driver transistors in the right channel. I think all the rest is an original full complementary configuration?
• Readings with a Fluke 87 after cleaning each of the circuit contact points I used for the readings.

Left Channel 1 driver reads 190 ohms. Outputs read 1.5 ohms with one at .9, so I am assuming bad.
Left Channel 2 driver reads 11.2 rest read 9.4

Right Channel 1 and 2 all transistors read in the 2+ mega ohm range.

• I am stuck trying to figure out this high reading in the right channel. All right channel emitter resistors read .33 or close. The thermal relay and all grounding points to chassis all read about .3 ohms. Every collector plate reads .3 ohms to the collector rail. I pulled the caps in the right channel and they read correct. The only difference I could find on the right and left paired transistor boards themselves is that across the collector rails equals 10 ohms on the left and 1.2 megohms on the right. The 5W R1 resistor on the control board is bad, but if I jumper across it the readings on the transistors do not change. All diodes on the control board
• The components shown on the main PCB layout Diagram 2.2 in the service manual are not present in this board.

I’m stuck. The flyback diodes are identified as d115-116 on page 20 of the service manual, but I cannot find them anywhere on the schematic from the service manual scan. I don’t have enough circuit analysis knowledge to identify them, but all diodes on the main PCB read .6 - .7nn with the Fluke diode test.

Thanks in advance for the help. I haven’t decided what to do with this amp yet, but regardless I like to fix things before making any mods.

 

[laatsch55]

The fluback diodes are on the backwall admist all those emitter resistors.
The 10ohms/ 1.2 megohms--- sounds like an open bias resistor.
Pull th outputs.for an out of circuit test. .

 

[laatsch55]

Dont see any diodes on the backwall...no biggie.

Full comp yessir..

 

[Gepetto]

Dont see any diodes on the backwall...no biggie.

Full comp yessir..

Cost reduction Lee...

 

[laatsch55]

That made someone a BIG bonus....

 

[Gepetto]

That made someone a BIG bonus....

Yeah the bean counter that asked for the cost reduction...

 

[wattsabundant]

There are shorted output transistors. The 700II manual is attached. Read the troubleshooting section for checking transistors and do exactly what it says. Note that the manual was written before digital meters were available.

 

[Mbose]

Thank you @laatsch55, @Gepetto, & @wattsabundant.

If I understood your comments:

1. There are no flyback diodes. They eliminated them to save cost.

2. As I thought, there is at least one bad transformer in the Left Channel. (I will pull them next, was more worried about the right channel's high resistance readings.

3. Thank you for what looks like a better copy of the service manual than the one I was using. I already ran the service manual tests before the first post, using both a Fluke and Simpson 260. That is what revealed the very high resistance in the right channel. Every resistor, capacitor, and even the thermal relay on the back wall tested ok. The backwall solder joints all seem ok as well.

Based on your comments, I just pulled and measured the right channel transistors with a Peak DCA75. Not all measured great, but they all did pass.

Question: I am assuming that regardless of whether the output transistors are ok, this cannot be powered up with an unknown source of very high resistance in the right channel? If that is correct, is it reasonable to assume that somewhere in the primary PCB there is either a bad solder joint or component in the right channel that is causing the high resistance? I have checked the diodes and resistors on the board, but will move on to the rest of the components if that seems reasonable.

Thank you again for the help.

 

[wattsabundant]

I don't understand what the high resistance is. The most common failure mode of transistors is zero ohms across one or both the junctions. In other words, if it's not shorted, it's good. Take that approach first. On the board, R101, 7.5k/2w has shown failures in the past.

 

[TimD]

I'm still trying to figure out what test points he is reading across . . . reread as many times as I can, and I still never see it mentioned, so cannot possibly make an observation.

As noted, good semiconductors should show conductivity in one direction and not the other across the junctions (but the meter test voktage has to be high enough for the junction under test to conduct). Caps typically start low and rise . . . Resistors should read as marked, and output/driver transistors on the heatsink should have infinite resistance to ground. Transformer windings will read very low, since the meter is dc, so they look like wire. Anything else, without context, is undefinable . . .

 

[Mbose]

@TimD I obviously have not done a good job of explaining this. I apologize for frustrating anyone and will try to be brief and clear.

The Service Manual says do not power up if fuses are blown. It says to check for shorted outputs. Before my first post I exactly followed the instructions and found the readings I reported in the first post.


Service Manual page 20:



This is what I measured. From the collector bus on one row of transistors, to any emitter, and then to each emitter. According to the Service Manual, if no transistors are shorted, the reading should be what's expected from the forward bias on the (imaginary) flyback diodes. As confirmed by @laatsch55 and @Gepetto (thank you!), there are no flyback diodes, so I don't know what the reading from the bus to an emitter should be if the transistors are good. I have no doubt what the reading should be if one or more transistors are bad.

On the left channel I got readings that showed at least one transistor was bad. One the right channel I got infinite ohms.
Bus to emitter on the right channel with good transistors is 10 ohms. On the left channel with good transistors it's mega ohms. Something in the right channel other than the output transistors is wrong and that is why I eventually was testing across every resistor, diode, and solder joint without disassembling the boards. I don't know if there are specific places where the 700s tend to fail, and I was hoping to get lucky with a shortcut to finding the problem. For example on a Yamaha C4 I would tell someone to first check the connection between the boards because those solder joints are terrible. Thank you @wattsabundant for the R101 tip.

At this point I just going to pull everything from the main PCB and test it, and reflow the solder joints. Again, I apologize for causing any frustration.

 

[TimD]

If there are no diodes, and no short, the resistance will be very high, indicating that that channel is good. Thus, on those tests, the channel with the *low* resistance has the fault.

 

[Mbose]

If there are no diodes, and no short, the resistance will be very high, indicating that that channel is good. Thus, on those tests, the channel with the *low* resistance has the fault.

That's very interesting. Will check it out tomorrow. Thank you!

 

[wattsabundant]

Is this amplifier quasi complimentary or full complimentary? Quasi complimentary will have all NPN outputs such as MJ15024, TP9054, or 2SD555. Full complimentary has PNP and NPN outputs. The manual was written for quasi-comp. That could explain differences that conflict with the manual. Also, digital meters with the diode test function didn't exist either. If a digital meter is used today, the resistance scale will not show a 0.6 volt drop across a P/N junction.