Phase Linear 700 Series II - restoration guidelines
Dear All Phase Linear Friends,
I probably won't be very 'eccentric' to say that Audiokarma is just a perfect place to find so many great solutions while repairing our valuable, certainly in terms of emotions, even most rare audio systems. I’m a strong believer in Phase Linear (P/L) heritage in power and pre-amplification. The brand history speaks by itself... Still if someone who lives far from any experienced P/L service (my case) wants to perform some required renovation works for the 700 Series II amp ( i.e. is able to do it), many important and undocumented tweaks are quite difficult to find even here. They are well spread around and despite educational benefits in looking for them in all various threads, it takes long time to identify the key ones.
That is why I thought to put together some major recommendations based on numerous P/L related AK threads, most often coming from D3imlay – many thanks Don!
Main purpose of the list below is to help in planning and performing the restoration of 30+ years old P/L 700 series II amplifier, and certainly not to re-engineer the gear. Replacing a vast number of components, that were working well so far and are not specifically exposed to aging, is not recommended. No doubt the temptation to try some great advice is in everyone of us – after all you decide at your own risk …
1)
GENERAL RULE: Use extreme caution while performing any replacements/upgrades due hazardous voltage presence in P/L 700 II power amplifier. Be sure that before starting any replacements the line cord is unplugged and power supply caps are discharged. Strictly follow ALL check-up and repair procedures as written in P/L 700 II Service Manual and subsequent service bulletins. Wear safety glasses while taking data from powered circuits and soldering. Before you start it is mandatory to have required equipment by the factory service manual (covered in later posts) and necessary tools - for toolkit recommendations see this thread for help:
http://www.audiokarma.org/forums/showthread.php?t=187673.
2) For powering up, before the check-up procedure starts, or/and for performed modifications testing, use light bulb test set (
http://www.audiokarma.org/forums/showthread.php?t=187673). Start with a 40 or 60 Watt bulb. Step up to 100 Watt or 200 Watt as your confidence builds. You can't have a load connected when you use the bulb. The amp puts out several volts DC as it starts up. If a load is attached, the bulb limits the current, as it is supposed to, and the AC input never gets above 30-40 volts. Use of variacs only will not limit the current and may burn all the output transistors (painful when you just replaced them…). If you still want to use the variac only, change AGC fuses to much smaller values to control excessive current.
3) Outputs and drivers should be reinstalled even if not destroyed. The reason are rubber insulators causing failures in time when overheated. If the amp had silicon pads on the outputs always get new ones - never reuse them! Don't over-tighten while mounting transistors. If you use mica insulators as preference, go easy with term compound grease. Keep it out of the sockets. Heat sink plates must be well cleaned before re-installation of outputs and drivers.
4) Following P/L service bulletins, output transistors in each channel should ALL be of the same type (e.g. MJ15024 or MJ21196 for quasi-comp. stage, or MJ15024/25 / MJ21196/95 respectively for full comp. stage). MJ211** are probably today’s most advanced high power TO-3 outputs and have higher SOA than MJ150**. In 700 Series II full complementary amp, the drivers and output transistors are the same. For example – the driver for the MJ15024 output transistor is a MJ15024. Same comment for the MJ15025 (PNP match for the MJ15024). In other words DO NOT mix different types of output transistors in a single channel. The previous statement is very important when doing service or repair work – not only with P/L equipment, but any kind / brand of amplifier. In 700 Series II quasi complementary amp there are obsolete RCA410 as drivers. Leave them, if not shorted, for better stability and sound (subjective impression). Last but not least, avoid buying the output transistors from unknown sources, counterfeited ones are often found. They won’t be of any use in 700 Series II severe working conditions.
5) P/L 700 Series II driver board (PL36) requires some closer attention. There are components that over last decades could loose resistance /capacitance desired characteristics or get open, even impacting other elements, not to mention about shorted diodes or signal transistors. To simplify our list, it is assumed that the PL36 and PL37 (display) boards worked OK before the restoration works had started and only outputs were actually replaced.
6) The resistors to consider for replacement, as they may drift from heat and age, are:
-> R1, R2 = 7.5K/5W 5% (these dropper resistors get so hot to even discolor the PCB card and affect nearby parts as well),
-> R101 = 7.5K/2W 5% (applicable per other channel too),
-> R103 = 2.7K/0.5W 5% (applicable per other channel too),
-> all power outputs emitter resistors = 0,33Ohm/2W (if you replace them, pick high grade with 1% tolerance 3-5W, coated wirewounds are nice).
7) In order to cure possible oscillations, usually after you replaced power outputs, the capacitors a 22~47pF usually work well to damp out any such tendency (see specific P/L service bulletin on the issue). The factory recommended 150~270pF value might be too high, and can cause high frequency slewing at high output power. Ceramic or silver mica types are recommended with at least a 100-Volt rating.
8) Replacement of all electrolytic caps on the driver board is highly advisable. Try to stick to high class brands designed for audio applications (e.g. Panasonic FC, Nichicom KZ, Rubycon ZA, ELNA Silmic II) - subject for sound tests and above all own music preferences. This upgrade most probably influences how your amp will eventually sound like. Some P/L fans say that ELNA caps are slowing and soften the music you hear, so good to listen for instance to mellow jazz. Sometimes not desired when you like 'rocking the house'… My own choice were Nichicon Muse and Rubycon high grade, but the latter are hard to find, especially for 100V (watch out for fakes!). Interesting comments on recapping can be found here by Arkay, see post #4:
http://www.audiokarma.org/forums/showthread.php?t=287093. Very informative post #15 by Echowars on matching caps' voltage is here:
http://www.audiokarma.org/forums/showthread.php?p=2637358#post2637358. No need to exaggerate though - please bear in mind that Phase Linear power amps are direct coupled and have no electrolytic caps in the signal path. So the replacement is a rather a precaution to avoid caps aging problem in the power supply, feedback network and bootstrap circuit (last remarks thanks to Dean & Ed).
9) The Op-Amp IC LF356 replacement with LF351 is recommended by the factory, so is the replacement of R1, R2 from 2.4K/5W to 7.5K/5W (the LF351 op-amp has better specs and helps with turn on/off thump). I don’t want to open a separate debate on whether LF351 is fine for front-end applications or not, in view of many options available these days. The op-amp in the P/L design had one purpose: to provide high gain and it was used to replace a traditional differential input pair - much more refined solution known from P/L 400-700 Series I. This had few advantages back in the 70's: all the transistors inside the op-amp could be closely matched (important for a differential amp) and production / testing was more efficient since matching of transistors was no longer necessary. LF351 is not at all a bad op-amp. So, let’s simply leave it as is. However, for All who like experiments it is highly recommended to put on IC 8-pin socket. Remember - the traces on the board are aged and therefore very delicate. The use of BB OPA134 op-amp as direct replacement to LF351 might be sound quality beneficial upgrade. For the music I personally like, and my whole audio setup, they work just fine – not too much ‘dark’ as some LM or AD IC’s enthusiasts may say… Besides for the other IC units you would need to make some power supply improvements (e.g. constant current source, higher voltage supply: >15V) and we were not here to re-engineer the P/L great amp, right?
10) Regardless whether you stay with LF351 or not, there are some common practices you may want to consider for the op-amp working conditions improvement. First is to add decoupling 0,1uF (e.g. per 100V) high grade film type cap on the +/- 15V rail supply and mounted directly on the supply pins for each op-amp (diagram attached to the thread #74, the solution helps in removing power supply noise and possible distortions at high frequency). Second would be to add 4,7uF/35V electrolytic cap ('reservoir') in parallel to each D1, D2 Zener diode. Be careful though - too much capacitance increases the turn-on thump.
11) Power caps are absolutely worth of replacing. They also strongly influence the sound when got dried over the years. Sometimes they may well become a source of hum. With the new ones go for twice the capacity than original - ca. 20.000uF @100V at minimum - you can find per 160V at small price difference = better (The originals were 9800uf @100V and they run right around 100V. With caps you buy today you don't get the headroom like you did 20 years ago...). The power supply caps testing: check if they have less than a few hundred mV of ripple at idle, at full load it will be slightly over 1V. Don’t buy power caps older than 3-5 years of shelf storage, if you don’t have time for reformatting them. Watch the cap size to avoid additional mounting problems. Before you begin, remember to discharge the power caps (a resistor 10-25 Ohm/10W would help)!
12) If you increase the power cap capacitance from original value of 9.800uF to say 20.000uF, you need to take into account the higher stress on your bridge rectifier. Some P/L owners decided therefore to upgrade it for >35A/400V units. In addition bypassing each diode by 0.01uF/1000V cap (ceramic or mylar HQ) will help to cure against RF unwanted interferences. BTW, you need to check the presence of C112=0.22uF/400V as no longer needed. Bypassing filter caps with two mylar caps 0.1-1uF/600V can be considered if you go for larger power capacitors. It will improve the ESR at higher frequencies. For 20.000uF cap, with low ESR, ‘computer high grade’ type, etc., etc. - you may not bother, probably.
13) Final review of the PCB card from the traces side is maybe a simple advice, but very important at the same time. Apparently over time the PCB material gets more ‘fatigue’ and you need to resolder anything that looks questionable. Cold/open solders are most time consuming cause for the amp malfunctions to find.
14) In the end bias setting is always needed. See related procedure in the P/L 700 Series II service manual. Here is a short wrap-up - not to forget about the step. So, when all the changes to your amp are finally done, use the bulb testing device to start the amp with no I/O connections. When successful, use a voltmeter to measure Voltage across the zener diodes. You should see +/- 15 VDC. Typical is around 14.5 VDC. Then with full input voltage check for about 300~350 mV DC on the 10 Ohm/0.5W resistors on the output buss (R109). Adjust the pot (R107) for the corresponding channel as needed. Recheck in ten minutes. Verify if less than 30 mV.
Be careful when taking data. A slip of the probe will cause major damage. Insulate the probe tip with tape or heatshrink. Always wear safety glasses.
15) When you performed all the upgrades, installing the output protection circuit is to be seriously considered to eventually save your precious loudspeakers... There are few options for P/L 700 Series II to follow and at AK you can easily find them (D3imlay). Otherwise just Google the topic, and for EU P/L amp owners there is a module I know 'made in the Netherlands', but also for instance another fine one from Tasmania (...). But one common mistake here is to forget about reconnecting the RF circuit (C108 & R135) from speakers’ terminals to the protection unit inputs. You will experience oscillation problems if you forget the precaution. Very useful educational case story for deploying the output protection for P/L 400 amp is found here:
http://www.flickr.com/photos/watts_abundant/sets/72157604339152710/ (by Don Imlay).
16) To conclude, a piece of electromechanical advice: replace all RCA inputs and speakers posts for something 'more sound neutral' than it is today. A variety of solutions can be found in other AK threads. Also don’t forget about the power cord – better to replace the old one if suspected condition. It should be non-polarized and not wall socket grounded (even for EU standards) – otherwise very likely hum problems appear. Keep the 'earth ground' connected closely to your source (CD/DAC/Pre-amp).
I refrain on purpose from touching the amp rewiring topic in more details. There is an issue here, known to all P/L users. In P/L 700 Series II amp the RCA input ground is correctly tied to the input ground of the amplifier's input circuits. Then the right channel's speaker ground is directly tied to the RCA ground... So ground current flowing into the right channel speaker ground is injected into the 'quiet' input ground (unwanted loop deteriorating the sound). Grounding is critical to both stability, measured performance and the amp sound quality. Star topology should be observed. At least one ground wiring should be added between a solid copper bare inside your amp and the common point on the power supply caps. Again, see Don valuable recommendations on that.
I welcome all additional remarks and corrections - nobody is perfect ;-)...
Good luck with your projects!
May Force Be With You....