February 22, 2004

Revisions since December 3, 2001

I finally decided to bypass the tone controls, having found a scheme that works.  And it does.  The instructions are up on Curcio's website, but are basically as follows:

1. Install a jumper between linestage PC board terminals 4 and 6.

2. Install another jumper between linestage PC board terminals 6 and 7.  This can be a continuation of the jumper in step 1 above.

3. Desolder and lift the wire connected to linestage PC board terminal 3 and insulate it.

4. Repeat the above three steps for the right channel using the corresponding (equivalent) linestage PC board terminals.

Results?  The preamp's personality changes dramatically and suddenly becomes sweet and totally unannoying.

Sorry it took this long!

December 3, 2001

Revisions since March 24, 2000

Guess I've been asleep at the old documentation wheel, so to speak.  Seems like I've missed some of the additions I made, particularly those in the power supply of the PAS since my last update.  I've been finished and happy with it for a long time now (for me), so let me see what I've missed and stick it in here.

1. I went with 60-volt Schottky diodes in the filament supply. Since then, I've used them in a Fisher X-202B as well as some Scott and Sherwood amplifier bias supplies. I still bypass them each with a .01mfd@500vdc ceramic "snubber" cap to reduce EMI/RFI, and haven't had any problems. Just watch your current requirements. Mouser has them in (I think) 3A versions at this voltage.

2. The other pertinent observation is that even though I have tried, I cannot find any tubes that are quite the equal of the original Telefunken 12AX7's that Dynaco supplied with the PAS. I have tried JJ/Teslas (which can actually brighten up the phono stage), JAN/Philips, Mullards and Ei's. Your mileage may vary, but the Tele's are King. No surprise there, but it makes a discernible difference and can change your opinion of the preamp.

3. If I haven't mentioned it previously, I also replaced the original can-type electrolytic cap with a currently available 20/20/20/20mfd@450vdc  multi-section cap and added a 33mfd@350vdc Xicon electrolytic cap to the B+ connection point of each circuit board--the phono stage and the linestage, so they each see 20 + 33mfd, or 53mfd of filtering.  I believe I also had to add an additional cap to one other section of the new can cap and might have paralleled two sections of the cap to make all of the values come out close to the original.  I'll fill you in when I get the thing opened up again.

It's not a bad preamp. Some crosstalk, kind of a lifeless phono stage, but not bad. Truth be told, however, I've recently rebuilt an EICO HF-85 preamp and like it better. Again, I used an oil-in-paper cap in parallel with a 716P Orange Drop in each channel of the output stage (as well as the tape outs) and it made a big difference. Sue me, I cannot tell a lie--it's better.

March 24, 2000

Revisions since June 23, 1999

1. I added a single .12 mfd @ 600vdc Sprague Vitamin-Q oil cap across each .22 mfd @ 630vdc 716P Orange Drop output coupling cap just to see what it would sound like. In other words, the Vitamin-Q is in parallel with the Orange Drop and increases the output coupling capacitance from .22 mfd to .34 mfd in each channel. More importantly, it smooths out the sound, making it almost liquid. I left them in, since the pc board is getting marginal at this point, and any more manglement would be criminal. Even driving my Hafler DH-500 solid-state amp, it's pleasant yet still revealing--a nice balance.

2. An upcoming test will be to replace the filament supply rectifiers--currently 1N4003's I think--with a couple of FR-whatever fast recovery guys I got from AES.

June 23, 1999

Revisions since June 14, 1999

1. You have to change the so-called dropping resistors (described below) in the redesigned filament supply from 1 ohm to 5-ohm each--I used a pair of 10-ohm 2-watters in parallel for each, as that's what Radio Shack stocks. (Dale resistors--not bad.)  Then you get about 11 volts dc on each PC board, or roughly 5.5vdc per 12AX7, which given that it's dc, is just about perfect. All this because of the increased filament supply output voltage due to the replacement of the dual selenium rectifier with silicon diodes. As I've said before, eventually I'll try Schottky diodes in this application, but for now the thing is sounding pretty good to me. It's a little amazing that they used half-wave recitfiers in this application but then again, the Marantz 7 used a half-wave silicon rectifier in the main B+ supply--go figure!

June 14, 1999

Revisions since June 6, 1999

1. Well, mea culpa, as they say. I FIXED THE HUM! It was a ground loop, and here's the fix--REMOVE the two .01mfd @ 2KV AC line bypass caps on the three-wire line cord (if you put them in as directed below). Then simply lift the Green wire from the chassis, and Presto--no more hum problem.  The short story is that if you just leave the original line cord alone, you will never see this problem. I use three-wire line cords on ALL of my equipment, and that includes the Stereo 70 I'm using with the PAS described here. I didn't try lifting the Stereo 70's ground, or plugging it into the PAS--I don't like passing all that current through the preamp's chassis.

2. As I had gone to all the trouble to put in the three-wire line cord, and as this hum problem only seemed to manifest itself when the PAS was used with the Stereo 70--not with a Hafler 500, etc.--I decided to put in a Ground Lift switch on the back panel. As my PAS came with two smashed AC line sockets--the outermost one and the innermost one--I simply put a SPST slide switch (from Radio Shack, of course) in the hole vacated by the socket closest to the AC line cord. Now when the switch is to the left, the ground is lifted, and when it is moved over to the right--the "ON" position--the chassis is tied to AC ground. The green wire now goes to one terminal of the switch, and the other terminal is wired to one of the power transformer mounting bolts. Cool, if I do say so myself. So now it's perfectly quiet, and capable of being grounded in applocations where it doesn't induce hum. NOTE: I still observe the National Electrical Code standard that says that the white (neutral) wire should go to the unswitched transformer primary, and the black (hot) wire should go to the Power switch--preferably through a fuse, which I added as described below. I probably have the fuse on the wrong side of the transformer primary, but any fuse is better than no fuse.

June 6, 1999

Revisions since June 2, 1999:

1. I heard a slight hum whenever the PAS was plugged into the Stereo 70 when connected to the Electro-Voice Marquis horn speakers, which are highly efficient. So I went back in and pulled out the extra 33mfd @ 450vdc electrolytic caps I had put in on the B+ attachment points of each circuit board (PC-5 and PC-6), and instead added a single 10mfd @ 350vdc cap in parallel with the 4th 20mfd section of the 20/20/20/20 mfd can cap I had installed instead of the stock (30-year-old) 20/20/10/30 can cap that was in there originally.  So now the four sections look like 20/20/20/30 mfd, which is pretty  close to original. I was worried that I had set up a ground loop the way I installed the extra caps, thus contributing to the hum.  It is VERY slight and although now reduced, is still slightly evident, but only when the roomis quiet and there is no program material playing. Mysteriously, it occurs with the PAS turned on or off, with the volume up or down, with the ground pin lifted or not, and with the lifted plug reversed or not. It only goes away if the PAS is either disconnected from the input of the Stereo 70, or if it is still connected, its power cord is unplugged. It also occurs if the Stereo 70 ground pin is lifted or not or if its plug is reversed or not. Go figure. I'll catch it someday. NOTE: all the equipment around here has three-wire line cords (with a few exceptions), and this is the first time I've ever experienced a ground loop problem of any kind--if that's the problem.

2. I also replaced the original 1mfd @ 25vdc electrolytic caps that are part of the -3X bass control circuit with (what I had) Solens 1mfd @ 630vdc FastCaps. I must say that it seems to have significantly smoothed out the bass--didn't eliminate the hum, though (which I thought it might).

June 2, 1999

Revisions since May 2, 1999:

Changes I've made in the filament supply have made a BIG difference in the characteristic sound of the PAS. Since replacing the selenium rectifier, I've felt that something I never liked about the PAS had crept back into the sound, so today I decided to implement something I had been testing on the bench a month or so ago--a second filter section in the DC filament supply.

So I added a (now) third 5-tab terminal strip close to the transformer so I could build a two-section filter in both the positive and negative 12vdc legs of the filament supply. Now instead of a half-wave diode rectifier feeding a single 4700mfd @ 35vdc electrolytic cap, each leg NOW features a half-wave 1N4004 diode rectifier (bypassed with a .01mfd @ 500vdc ceramic cap), feeding a 2200mfd @ 25vdc electrolytic cap, a 1-ohm, 10-watt series "dropping" resistor, and a 4700 mfd @ 35vdc electrolytic cap on the output. On the scope, the pronounced triangular waveform seen on the original design after replacing the selenium recitifer with diodes is now much smoother, like a slightly tilted mound (or maybe an Almond Joy bar). The output voltage is down--not a lot, but down--from 12.6vdc to about 12.2 vdc on each side. And best of all, that annoying peak that aggravates you when you turn it up is nearly gone. I guess Schottky diodes are next. Update--use 5 ohm 5-watt resistors instead of 1 ohm resistors (see the June 23 revision above), which then yields 11vdc on each side.

So, now my feeling is that it's probably best to leave the selenium rectifier alone unless you're going to do some serious re-engineering on the filament supply. I guess this is common knowledge, but I had to learn it the hard way. I don't know if a regulator on each leg would help or hurt at this point, but I have one of each--a +12 and a -12 3-terminal regulator, waiting to be tested, so I'll find out sooner or later. Let me say, though, that this present change is major. And I mean that in a positive sense.

May 2, 1999

Revisions since April 30, 1999:

1. Remembered that I put a fuse in the damned thing when I installed the three-wire line cord, so I thought I'd better document it (see below).

April 30, 1999

Revisions since April 14, 1999:

1. Replaced the can cap in the power supply with a 20/20/20/20mfd@475vdc can cap I took out of the Stereo 70 I rebuilt, and added a 33mfd@450vdc axial cap under the chassis on the B+ attachment point of EACH PC board.

2. Tried and then removed a 1mfd@630vdc Solens Fastcap bypass on each additional 33mfd@450vdc axial power supply filtering cap described in #1 above. Contributed to a midrange peakiness, I felt--subject to reconsideration.

3. Removed 1mfd@200vdc film cap at each bass control to linestage PC board and replaced the original 1mfd@25vdc electolytic caps at bass controls. May also have contributed to midrange peakiness--subject to further testing with better caps (like the Solens' I removed in #2 above).

4. Replaced small-value caps on linestage board with new silver micas.

5. Stuck a pair of used 12AX7 Amperex Bugle Boys inthe linestage board--nice clean highs and decent bass. Quiet, too. Very crystalline.

April 14, 1999

Welcome to my mod page for the Dynaco PAS2/PAS3/PAS-2X/PAS-3X preamp. The PAS is a active preamp (now that there are such things as passive preamps) comprised of four 12AX7's with dc on their filaments and a 12X4 tube-based full-wave rectifier. Interestingly, the preamp, consisting of two circuit boards--a phono preamp and a linestage--uses two 12AX7s per board, employing series paralleled filaments on each, and a Plus 11-volt dc supply for one board and a Minus 11-volt dc supply for the other, with a selenium full-wave rectifer and two 2200mfd@18vdc electrolytic caps in the filament supply.  There is no hum balance control, but the preamp is essentially totally devoid of hum.  As is typical with 30-year old plus designs, it has magnetic phono input, a high-gain tape head input, a "Special" input configured as a microphone input on factory-wired versions, and three line-level inputs--AM-FM, FM MPX, and "Spare".  It also has a tape recording and monitoring loop, a high (scratch) filter, as well as a Loudness switch.

Of course, a schematic is always helpful.  Fortunately, there is a complete manual for the PAS-2 up on Curcio Audio's website, which you can download or read in Adobe Acrobat by clicking here. For a full manual copy of this and other Dynaco products, Sound Valves sells--and occasionally auctions off at eBay , which is where I grabbed them--two different collections in binders, one for Preamps/Tuners and one for Amps for $49 each.  I got them both, and I'm glad I did.

My Personal PAS Story

I built a PAS-3X in my dorm room at the University of Rochester back in 1970 or so and used it with a pair Mark III's I built at the same time (still have them). Mark III's were $65 mailorder kits back then. The thing about the PAS is that I just never really liked it for some undefinable reason. I suffered through innumerable tube changes which later proved to be more a result of my stupidity and lack of technical know-how than a real need (see below). I also saw Dyna struggle to maintain a decent supply of replacement tubes, as they went from OEM Telefunkens to India-sourced Mullard knockoffs, etc. I eventually sold it and got a CJ PV2-aR, which I still have and love.

Then recently a guy (Thanks, Paul!) sent me a factory-wired PAS-2 with what turned out to be the -3 faceplate conversion with some wacked-out knobs--an FM-3 tuner tuning knob for the volume and an SCA-35 small knob for the balance knob. It also had the PAS-3X tone control and blend control upgrade installed. Real replacement PAS-3X tone controls--the patented variety--are now completely unobtainable, by the way. The -3X patented pots--both bass and treble--are identifiable from the rear by their two lubrication/cleaning ports (or holes, really), as opposed to the non-3X tone controls, which only have the standard one cleaning port (hole) on the body of the pot. This is clear in the rear view of the assembled front panel in the manual. Also, the unit I received has the 1mfd electrolytic caps from the bass controls to the linestage board, as was incorporated in the -3X design. The tone control mod and changed blend control resistor values was in an update kit which is the subject of a Dyna Service Bulletin I found in SoundValves' manual binder for Preamps and Tuners. Their Amp binder is good, too, and both are must-haves if you work on this stuff.

So I started by listening to the bone-stock unit (still with its original Telefunken Dyna 12AX7's). It had a bad 12X4, but once replaced, the unit played and sounded like I remembered it--somewhat congested highs, an bit of an annoying midrange, but an overall buttery sound.

The PAS uses a selenium rectifier in its dc filament supply, and as these type of rectifiers exhibit increased series resistance over time (as well as emit toxic fumes if you incinerate them), the filament voltage on a vintage PAS is on the order of 11 volts dc or less, and as such original tubes tend to last a LONG time (reminding me of my original stupidity mentioned in the first paragraph). In fact, savvy buyers snap up original PAS' just to get the tubes, as they are easily worth around $160 for the quad, used.

This first treatment explores the potential of the PAS with it's stock circuit.  Future updates will explore modifications such as tone control circuit bypassing, regulated filament supplies, etc.

Replace All of the Caps:

To start out, more as a feasibility test than anything, I replaced the two blue .22mfd@400vdc output caps (those ugly and notoriously grody blue blobs) with 1.0mfd@400v ECI polyester film caps from AES, and heard something I hadn't heard in all the years I listened to my original PAS--music and a future. I then replaced ALL the caps in the PAS--except the 2% mica wafer-types used in the RIAA circuit--with either silver mica (for the four small-value caps on the linestage board) or with original value Orange Drops--716P 400vdc units where I could get the original values, and PS-series where I needed specials (such as .0075mfd and .004mfd for the tone control circuits)--for everything else. I also replaced the selenium rectifier with a pair of 1N4004's each bypassed by a .01@500vdc ceramic cap, and replaced the 34-year-old 2200mfd@18vdc caps with a pair of 4700mfd@16vdc electrolytics. Also out were the 1.0mfd electrolytics from the bass controls, which were replaced with Radio Shack 1.0mfd@200vdc film caps. Update--after much thinking and listening and looking at the circuit, I decided to put the 1mfd@25vdc electrolytics back in the tone control circuit. Also I removed a pair of Solens 1mfd@630vdc Fastcaps I had bypassed the B+ to each PC board with, thus entirely removing an annoying midrange peakiness I had somehow introduced into the design.

I also cleaned all the pots, switch sections and tube sockets with DeOxit5 deoxidation spray solution (another must-have).

I then shuttled the tubes around until I got the sound I liked (still using the stock tubes--34 years old, mind you), and I can now honestly say that this preamp does not take a back seat to anything out there for under a grand, and my total investment is maybe $25 plus the PAS. It absolutely rocks and rolls. AND is dead quiet. (I haven't listened to the phono stage yet.) Update--I did drag an old Philips belt-drive with a Stanton 500 cartridge up to the bench and it sounded a little rolled-off--more to come on this part of the deal.

Rebuild the power supply:

One other thing I did that I ALWAYS do and nobody ever mentions is replace the line cord with a 3-wire, 16-gauge line cord (or larger--I used Radio Shack), bypassed with .01@2KV (line-rated) caps from each side to chassis ground. You can use Panasonic X-Y Series (best) or Radio Shack ceramic caps (also fine). This is a must-do, in my opinion. The IEC says green is chassis ground, so attach that to the chassis (e.g., transformer mounting bolt if not insulated--I used the filament supply cap stack mounting screw).  White is neutral, and goes to the unswitched side of the power transformer primary winding. In this case, that would normally be connected to one of the convenience outlets. Can you believe the PAS doesn't have a fuse? PUT IN A FUSE HOLDER! I used a one-bolt chassis-mount fuse socket (like you find inside equipment when they don't want you to change it too easily, either because they want the money to do it or because if it blew, something is seriously wrong). Into that I put a 1 1/2A fuse--it's what I had, but it should be smaller. Then connect the white wire to one side of the fuse holder, and the other side of the fuse holder to where the white wire would have gone, namely the convenience outlet that has the unswitched side of the transformer primary connected to it. Black is HOT, so this goes to the AC power (on/off) switch. Connect a .01@2KV from black to green and another from white to green at the point of entry of the line cord, and you're done. Not more shocks, no more hum, no more noise, and no more electrical code violations. UPDATE: See the June 14 revision above--the bypassed three-wire line cord mod seems to be a problem with some amps. Connect the white and black wires as described, but read the section above on the green wire.

I also replaced the can cap with a 20/20/20/20mfd@475vdc Aero job that I pulled from the Stereo 70 I rebuilt, and added a 33mfd@450vdc electrolytic to ground at the B+ attachment point of each PC board.  It looks like the original can cap was good, but I've beeen having so many problems with can caps recently that I just threw it in there. What the heck, and with the extra 33mfd caps, each board sees 57mfd. I tried bypassing each of these 57mfd values with a 1mfd@630vdc Solens Fastcap as an experiment, but I pulled them out, as they introduced a midrange peakiness into the design, proving that you can over-upgrade. Be careful of this tendency--more is not always better, despite the claims of the so-called experts. Fact--tone controls are not inherently bad-sounding. Fact--balance controls do not inherently degrade the sound. Fact--electrolytic caps are not the worst thing you can have in the signal path, although I try to eliminate them where possible.

Sit Back and Let 'er Rip:

When you think you're done, plug everything in and try some program material. Actually, it's smart to do this at each stage of mods, so you don't have a problem and lose track of what you just did, which probably caused it. This is a Work in Progress, by and large, but without getting into any major overhauls of the design of the preamp, this is probably where you can stop.

So get out there and mine this gold before the eBay tube thieves snatch them all up. Hope you like it--I do. Makes a great little system with a rebuilt ST-70, a Rotel 930 budget CD player, and a pair of Dynaco A-25 speakers. Have fun and let me know how you make out.

To get back to the WA3FLE homepage, click here.