Phase Linear 400 mkII refreshing

[Milos]

So I bought this one for purely sentimental reasons. I could have paid double for something in a perfect state but I opted for something that calls for some love and attention... at the end I'll know what's inside.

It works, sounds a bit flat on lower volumes, and throws two "thumps" after about 3 sec when powered off, only to the right channel. Inside is a spiderweb of hot glue threads and a pair of undistiguishable blue cans of 6.800mF, much smaller than originals. Even some hot glue mess on poles of those capacitors (whyyyy?!). I see hot glue around smaller capacitors on the board as well and like tiny brownish burn marks on the two 6-pin transistors/semi-conductors on the board, could be just burned dust. (pictures coming soon)

But what disturbs me is the layout of transistors and I need advice there from you guys (see the image)
Previous owner(s) probably used what was salvageable from the original PL909's and moved them all to the right, then replacing left two grids with 2SC1116 and BUX48A at the bottom.

I already exchanged some emails with Don from Watts Abundant and protection relay/board is on its way. I'm gonna get a pair of those 15k cans from WO but when I dive into the guts I'm not closing it before these transistors are sorted out. I need some guidance with them and resistors which I should use next to them. Electronics are not my field though I know my way around with soldering... but anything more complex that this, someone else has to do it. It is going to stay quasi for now, until time and budget opens up for the board and backplates.

Thoughts?

 

[laatsch55]

It appears the FPL-909's are being driven by FPL-909's, the other channel the 2SC1116' are being driven by the BUX40's?? I've never seen that combination of drivers and outputs.
You should consider MJ21196's as outputs and RCA-410's as drivers....

 

[Gepetto]

So I bought this one for purely sentimental reasons. I could have paid double for something in a perfect state but I opted for something that calls for some love and attention... at the end I'll know what's inside.

It works, sounds a bit flat on lower volumes, and throws two "thumps" after about 3 sec when powered off, only to the right channel. Inside is a spiderweb of hot glue threads and a pair of undistiguishable blue cans of 6.800mF, much smaller than originals. Even some hot glue mess on poles of those capacitors (whyyyy?!). I see hot glue around smaller capacitors on the board as well and like tiny brownish burn marks on the two 6-pin transistors/semi-conductors on the board, could be just burned dust. (pictures coming soon)

But what disturbs me is the layout of transistors and I need advice there from you guys (see the image)
Previous owner(s) probably used what was salvageable from the original PL909's and moved them all to the right, then replacing left two grids with 2SC1116 and BUX48A at the bottom.

I already exchanged some emails with Don from Watts Abundant and protection relay/board is on its way. I'm gonna get a pair of those 15k cans from WO but when I dive into the guts I'm not closing it before these transistors are sorted out. I need some guidance with them and resistors which I should use next to them. Electronics are not my field though I know my way around with soldering... but anything more complex that this, someone else has to do it. It is going to stay quasi for now, until time and budget opens up for the board and backplates.

Thoughts?

Milos, those 2SC1116 devices are 'no bueno' in this application. They only have a C-E rating of 120VDC and are being applied in a 160V+ application. You should be obtaining MJ21196 devices to replace the devices in the upper 3 rows.

 

[Milos]

It appears the FPL-909's are being driven by FPL-909's, the other channel the 2SC1116' are being driven by the BUX40's?? I've never seen that combination of drivers and outputs.
You should consider MJ21196's as outputs and RCA-410's as drivers....

Yup! I've been googling like crazy to find any connection between 2SC1116 and PL400 but nothing. I'm lacking tech backround but it felt wrong so I had to ask.

Milos, those 2SC1116 devices are 'no bueno' in this application.

Advice regarding MJ21196 and Phoenix forum are a common google search result as the most relevant in this application so I figured that out even before joining the forum, I was basically looking for some reassuring that current layout is wrong, so thank you both for confirming this.
OK, so Motorolas, 12 of them should be widely available... what about the RCA drivers? Who's treasuring those?

I also dug out this, which only confirms:

I believe the 2nd and 4th heatsink is the negative DC leg for each channel. In quasi comp systems it is critical that the RCA 410 is used, as opposed to an MJ15024/21196 to prevent oscillations. I learned the hard way that using the same transistor for driver and outputs leads to oscillations. I stare at the schematic and struggle to understand why. I spent hours trying slow down tricks, base stoppers etc and nothing worked.

 

[Milos]

I'm being silent as I'm in the process of sourcing components for my project... and along the way, digging the web for replacement drivers I came across these fellas. Littlediode sells them but I also found some at Radwell and they aren't cheap. (Not to mention that Radwell has a ripoff shipment rates, quoting me 99.00 euros for a small envelope from Belgium to The Netherlands, obviously a no-go).

Has anyone ever come across RCA410 like these? I thought they were last made in mid 80's... I'm yet to measure these and compare.

 

[wattsabundant]

The date code is 40th week of 2001. (0140) they appear to be legit RCA's. Ideally all drivers would be RCA410's. If you're buying transistors, pick up Berquist silpad insulators. No grease required.

 

[Milos]

The date code is 40th week of 2001. (0140) they appear to be legit RCA's. Ideally all drivers would be RCA410's. If you're buying transistors, pick up Berquist silpad insulators. No grease required.

Don,
Thanks for this clarification and the advice, I'll avoid the mess with a paste. I was suspicious to even see them being sold new but I gave it a go just to see what I'm gonna get in the package... picture on the selling site is just showing generic image of a transistor without any markings on it.
Do you think there's any difference between these from early 2000's and those made in 80's, with a classic RCA logo on them? (I managed to find another pair from '86)

 

[mlucitt]

I would be curious about the Gain (hFe) readings on these newer RCA transistors. Real RCA 410 (and the exact same RCA transistor with a different P/N 66546) measure anywhere from 14 to 44 on my Sencore TF-151A. Your readings will be different but the range should be similar. Newer transistors are much more consistent. The RCA410 or RCA66546 are used even today because there is not an equivalent slow, low gain transistor for the Driver position. This slower (really lower frequency bandwidth) helps to prevent oscillations in the amplifier. The Driver Transistor does not do much of the amplification, rather it sets up the Bias for the remaining Power Transistors, in my opinion.

Also be aware of the huge black market of fake semiconductor devices, especially from small online sources. It is extremely easy to wipe the markings from a transistor and add different markings. Phase Linear did this when they wiped the Delco automotive ignition device, P/N DTS411, and re-stamped the power transistor with their logo and P/N PL909 or XPL909 for audio use, with Delco's approval, I'm sure.

 

[Milos]

I would be curious about the Gain (hFe) readings on these newer RCA transistors. Real RCA 410 (and the exact same RCA transistor with a different P/N 66546) measure anywhere from 14 to 44 on my Sencore TF-151A. Your readings will be different but the range should be similar. Newer transistors are much more consistent. The RCA410 or RCA66546 are used even today because there is not an equivalent slow, low gain transistor for the Driver position. This slower (really lower frequency bandwidth) helps to prevent oscillations in the amplifier. The Driver Transistor does not do much of the amplification, rather it sets up the Bias for the remaining Power Transistors, in my opinion.

Also be aware of the huge black market of fake semiconductor devices, especially from small online sources. It is extremely easy to wipe the markings from a transistor and add different markings. Phase Linear did this when they wiped the Delco automotive ignition device, P/N DTS411, and re-stamped the power transistor with their logo and P/N PL909 or XPL909 for audio use, with Delco's approval, I'm sure.

I'm not gonna say I was hoping for this kind of a revelation but I'm glad you provided this insight. This is the other pair I got and that's it, got no more, either all 4 together are gonna do the magic or I'm back to beginning. Worst case scenario I'm gonna bite it and do the whole thing with White Oak boards and motorolas all the way.

Btw, I have no idea why I thought these were from '86... maybe the number got stuck in my head as I read somewhere they've been made up until 1986.

 

[mlucitt]

From the AI-sphere:
The RCA 410 transistor, developed in the early 1950s, was part of RCA's significant contributions to transistor technology. The DoD assigned NSN 5961-01-0006-8005 to this transistor in 1975.

Phase Linear probably bought a majority of these transistors during production of the PL400 and PL700B in the 1970's. Discontinued in 1982 (only because I can find no reference to the RCA410 after that date). Phase linear later used the updated P/N RCA66546 until the firm was purchased by Pioneer Electronics in 1979.