original '67 Princeton Reverb - repairs advice and bright switch

[thelonious]

A friend of mine got an excellent condition, one-owner '67 Princeton Reverb and an original Maestro Fuzz Tone from a family friend for $220. The amp had about $250-350 worth of vintage tubes in it, too. What a deal. The first time he played it (in bare feet, in a basement! - he didn't know any better), it shocked him badly. I'm planning to remove the "death cap", change to a 3-prong power cord, replace the cap can and bias circuit filter cap, and clean the pots. I read a few posts, including this one: http://www.el34world.com/Forum/index.php?topic=13534.0 that were helpful in determining what I should and shouldn't do. I know there are a lot of opinions about how much these amps should be "repaired" vs. left alone, and I don't want to mess it up or diminish its value. On the other hand, he's not a collector---he's a gigging guitarist who plans to use this thing for the next 20 years. I haven't opened it up yet; I'm just trying to form a plan and get some info first.

Other than the shocking issue and dirty/scratchy volume pot, his primary complaint is that he has to have the treble and bass knobs near max in order to get a nice clean, scooped sound. I'm thinking that the lack of bass issue is related to the bypass caps drying out over time. My plan is to replace one bypass cap at a time until we get closer to the amount of bass he's looking for. Does that seem like a good way to approach this?

Re the treble knob, I noticed that he's keeping his volume around 2-3 in his apartment and 4-5 when gigging (it gets a little farty and displays some ugly-ish clipping above 5). Especially at apartment volume, I thought a bright switch might really help him out, but it seems almost like blasphemy to put something non-original in this beauty. Could I completely disconnect the "ground" switch in the back (which I hear some guys change to a standby switch) and run some shielded cable and a bright cap from the volume pot to that in order to turn it into a bright switch for him? Would that diminish the resale value? It's a completely removable mod. The only difference would be modern solder vs. vintage solder at the volume pot solder joints. Is that a bad idea?

Any other advice?

Gratefully,

Concerned in Chicago  

[sluckey]

Could I completely disconnect the "ground" switch in the back (which I hear some guys change to a standby switch) and run some shielded cable and a bright cap from the volume pot to that in order to turn it into a bright switch for him?

I would advise against running that low level signal from the vol pot over to the power section even with shielded wire.

Just for a test, temporarily tack solder a 120pf across the vol pot high side and wiper. If he likes it consider leaving it like that or also consider replacing the vol pot with a pot that has a push/pull switch.

[HotBluePlates]

And as a previous owner of a number of Princetons of this vintage, I would advise the "death cap", 3-prong power cord, cap can and bias cap changes and pot cleaning.

Then listen to the amp again. If there is any other speaker/cabinet available, plug it into the amp in place of the internal speaker to verify you're not fighting a tired original speaker (I never had a problem in mine).

The circuit point where a Mid pot would be in this amp is a fixed 6.8k resistor (like having the Mid on a different blackface Fender on about 7). You may just need to reduce the value of that resistor to get more "scoop".

If you still need a bright cap, add it as Sluckey suggested, with a new push-pull pot (though this may/may not fit in the amp because of the slope of the faceplate).

"Lack of bass" is probably not the amp's fault, but what he's expecting to hear when playing alone. Guitarists (myself included) often play with waaaayyy too much bass when playing alone, because we expect a "full sound". In a band situation, that 250Hz and below comes from the singer and bass player. Now, the "full sound" playing alone at home equates to "mud" when playing with the band.

I'd argue that every recording you ever heard has a guitar sound *way* brighter than you think it is, if you ever heard the guitar on its own.

I say this cause my old '67 PRinceton Reverb had bypass caps that had dried up to around 1-5uF each, and it was still warm and round, and way too muddy for live playing without setting the bass to ~4 and cranking the treble to max. And that was years after I sold that amp before I ever got into a brighter guitar sound.

[thelonious]

Super helpful. Thanks guys. I'll let you know how it goes!

[PRR]

> 250Hz and below comes from the singer and bass player

With few exceptions, it's not the singer filling the <250Hz zone.

The USSR's Army Choir (1938) had some heavy bottom (I learned on a restoration project). Still falling below 200Hz.

Bing Crosby "White Christmas" has strong 55Hz and 110Hz, but not strong like bass or drums.

Johnny Cash "16 Tons" sounds low, but little below 105Hz. (It's implied bass, the overtones line-up on a pitch he doesn't really project.)

Zero Mostel comes to my mind, but few of you have been in the same room with him, and the available recordings are not the same.

You get the right guy, Il Commendatore in Mozart's Don Giovanni can really thunder. The guy gets paid well to do nothing for 2 hours, then one 5 minute part playing a zombie-statue who pulls Don down to Hell. But most guys can't carry a large hall with heavy bass, the orchestra doubles it.

Typical rock/country singing doesn't have near as much bottom as these examples.

Rod Stewart "Wake Up Maggie I..." -- the graph falls below 150Hz, and this includes back-up band (Rod's way out front, but not a capella).

[thelonious]

Now that I have this thing in the "shop" (i.e., my basement), here are gutshots. This thing is a beauty. Original Jensen speaker is in perfect condition, too. I notice that the faceplate says "design and circuits patented 68" - so I assume it is not actually a '67 as claimed by the original owner?

Cap can looks original (1966 date code). Bias cap and resistor were prob changed in late 70s (cap has a 1976 date code). Turns out this is an AA764 model, not an AA1164. If what I read on forums is true, the main difference in AA764 is the use of GZ34 instead of 5U4, and maybe a different PT.

A couple questions...

Should I change the rectifier to 5U4 to bring voltage down or leave it alone? It has 465V plates/455v screens. With those voltages, just about the only modern power tubes he could run in this thing are JJs. I suppose the 5U4 would make it just a little quieter and saggier.

Do the bypass caps look original? I thought usually the originals were big brown Mallory paper caps, but I'm pretty new at this stuff. The light blue caps are Ducati branded. I can't tell what brand the smaller, black, common-negative caps are, and I can't find date codes on any of the bypass caps.

[eleventeen]

Thing of beauty. A local school summer music camp got a BF one in in utterly museum condition. It was all I could do not to pick it up and run off with it. I was thinking of offering them three new $125. little SS amps for that old thing. One of the best amps, ever.

[HotBluePlates]

It looks 100% stock except for maybe the resistor on the bias board and a couple of wire splices (strangely, they use the same cloth-covered wire on both ends of the splice, making me wonder if it was factory).

The QF date code on the tube chart is June '67, which jives with the can cap date code of 52nd week of '66.

The Ducati single-25uF bypass caps are stock too; I had some in one of my blackface Princetons, and never could find info on a cap maker named Ducati (the tech I asked at the time only heard of motorcycles by that name).

I *think* I remember my '67 Princeton Reverb also saying "Design & Circuits Patented 68". The 68 may not refer to 1968. Or maybe it does, and they used the panel in 1967 and intended to continue use into 1968. However, in 1968 the front panel changed to the silverface panel, and had the aluminum drip-edge around the baffle board.

EDIT: You'll find scores of pictures of blackface Princetons that also say "68" on the faceplate. I noticed that even the reproduction faceplates saying "Fender Elec. Instr. Co." have the 68, and they used that face plate in either 1964 only, or only through early 1965.

I think you'll find the speaker's date code to come out to 67 or late 66, as well as pot and transformer codes. If so, I'd call it a definite 1967 amp.

[sluckey]

It looks 100% stock except for maybe the resistor on the bias board and a couple of wire splices (strangely, they use the same cloth-covered wire on both ends of the splice, making me wonder if it was factory).

I think those splices are factory too. Look at the number of wires coming out of that PT. Looks like a dual primary. And those splices 'may' be associated with making the jumper connections. Wire colors in the pic match the wire colors on this schematic...

http://ampwares.com/schematics/PRINCETON_REVERB_AB1270.pdf

I know this is a later schematic but the PT may be the same.

[clyde]

The patented # is definitely not the year, but might be a chronological designation by Fender.  My Vibroverb says "59", doubt there was any notion of such an amp in 1959.

[thelonious]

Repairs are all finished, and this thing sounds like a million bucks. The owner loves it. Thanks for your advice!

One interesting thing we noted is that NOS power tubes seem to make the tremolo really happy, but new JJs make it weak (the intensity at '10' with JJs is similar to intensity at '4' or so with the original tubes). Is this because JJs are different enough from a traditional 6V6 that they bias differently and mess up the trem? We'll probably just stick with NOS tubes anyway, but has anyone else run into this problem? He'd like the option to use modern power tubes, too, and now I'm thinking the only way to make this work would be to install adjustable bias.

[HotBluePlates]

... NOS power tubes seem to make the tremolo really happy, but new JJs make it weak (the intensity at '10' with JJs is similar to intensity at '4' or so with the original tubes). Is this because JJs are different enough ... that they bias differently and mess up the trem? ...

You know the answer must be "Yes" because the trem is responding differently.

However, this may not be a fundamental difference in the tubes, as much as it is the natural variance in how individual tubes of a given type bias up.

You probably bias by measuring current. If the transconductance (Gm) of two tubes is greatly different (expressed by Europeans as mA/V), each will need a  different bias voltage to idle at the same current. If Gm is very, very different then the bias voltage will be very, very different.

I'm also guessing you didn't measure the bias voltage along with the idle current when biasing each set of tubes. Go back and check that while biasing each pair to the same idle current. My guess is that a different (and larger) bias voltage is needed for the JJ's you have, and a bigger bias voltage makes the fixed trem output seem less effective.

If the RCA's bias up with less bias voltage needed, then the trem signal looks bigger compared to that bias when using these tubes, and thereby makes the trem seem to have more depth.

Maybe something else is going on, but largely the same thing occurs with 6L6's vs 6V6's, in that 6L6's generally use a larger bias voltage for an acceptable bias in the same amp, and result in the trem seeming weaker.

[sluckey]

One interesting thing we noted is that NOS power tubes seem to make the tremolo really happy, but new JJs make it weak

I noticed that exact situation with a recent build, Ampeg J12. I just chalked it up as a difference in tube biasing.

[thelonious]

I noticed that exact situation with a recent build, Ampeg J12. I just chalked it up as a difference in tube biasing.

Good to know, and glad I didn't just return the amp to him with something bogus going on in the trem circuit!

You probably bias by measuring current. If the transconductance (Gm) of two tubes is greatly different (expressed by Europeans as mA/V), each will need a  different bias voltage to idle at the same current. If Gm is very, very different then the bias voltage will be very, very different.

I'm almost certain Gm is different in the JJs. You guys have probably seen this comparison since it's been kicked around for a while, but just in case you haven't - http://billmaudio.com/wp/?page_id=370

Sheepish admission: because originally this is a non-adjustable bias amp (fixed 22k resistor, or in the case of later models, 27k?), I didn't go beyond checking that the bias voltage was close to spec. With the tubes we ended up selecting, it was -32v when the schem said -34, so I left well enough alone beyond that, given that our goal was to return to spec. My guess is that once he lives with it for a little while, he's going to ask me to switch it to adjustable bias.

Next time he brings it in, I'll measure all of that with the JJs, too, and compare.

Is switching these to adjustable bias common practice? Will it lower the value?

[HotBluePlates]

... because originally this is a non-adjustable bias amp (fixed 22k resistor, or in the case of later models, 27k?), I didn't go beyond checking that the bias voltage was close to spec. ...

Yeah, unfortunately, you'll probably have to use the OT Shunt method of plate current measurement to see where idle current is for this amp.

If you have a mutual conductance tube tester, you could measure each tube for its Gm to help figure out what's happening. But you can guess similar information by seeing idle current draw at the same bias voltage. Lower current for same bias probably equals lower Gm.

Is switching these to adjustable bias common practice? Will it lower the value?

Sometimes it's done. It won't lower the value if you do it properly.

By that I mean you could replace the 22kΩ on the bias board with a small pot and a resistor to ground. There are small multi-turn pots that will fit in  the space on that board. The amp will look completely stock, but just requires you to pull the chassis to adjust idle current.

If you drilled a hole to add a pot that can be accessed without pulling the chassis, yes, you'll reduce the amp's value. The chassis would still have to be pulled unless you added some means of measuring tube current externally, so modding to add a bias adjust that doesn't require the chassis to be pulled seems like a bad idea given this is a collectible amp.

[Willabe]

By that I mean you could replace the 22kΩ on the bias board with a small pot and a resistor to ground. There are small multi-turn pots that will fit in  the space on that board. The amp will look completely stock, but just requires you to pull the chassis to adjust idle current.

I bought 1 from Doug, he still sells them, works fine.


                Brad      

[HotBluePlates]

I bought 1 from Doug, he still sells them, works fine.

Yes, but there's not a convenient way to mount those.

I was thinking of the Bourns trimpot below. I've put those in an amp by standing it on the end that is opposite the screw, and holding it on the fiber board with a dot of glue. You want to mount your resistor and wire on the pot before placing it, as it will make wiring easier to deal with.

[thelonious]

Great idea. That looks like maybe one of the Bourns 3006 series - 3/4W multi-turn trimmer? I looked at Doug's http://www.el34world.com/charts/fenderservice5.htm page under "Rewiring the Bias System", and it seems pretty doable.

Thanks again to all for your help!

[Willabe]

Yes, but there's not a convenient way to mount those.

I just put 1 leg in the PR's bias board, works fine. It's so small that there's no leverage to twist them.

The possible problem with the Bourns pot's is rotation life time.


                 Brad     

[HotBluePlates]

Yeah, I was just looking at that. Never saw their specs before yesterday.

I wouldn't expect Doug's trimpot to be any better, because after all they're trimpots... they're meant to be set and left alone (hence the poor ratings for rotational life compared to panel-mounted pots).

I guess if life is an issue, you lash-up a small panel mount pot in the chassis for bias control. The real issue is always that you don't want to add any holes in the chassis, so it can always be returned to stock upon resale.

[sluckey]

I've tweaked hundreds of this style Bourns trimmer pot over the last 40 years. I've replaced 2 or 3 and I've tweaked maybe a dozen others that would have a rough (jittery) spot at some point of the rotation, but it wasn't the point I needed, so the pot was not replaced. Some of these pots were tweaked once and never touched again, but some were tweaked daily or weekly. Most were tweaked annually. All were in circuits that were under 60 volts. The pots were all mil spec and pricey. I've never been concerned with the 'rotation' life. I've used them twice in guitar amp bias circuits. Here's a pic of a Harvard amp with bias trim.

[Ed_Chambley]

On old collectable fenders I have removed the external speaker jack and put a cts trimmer pot there.  Easy to change back and use the bias checker that plugs between the tube and socket I can adjust without removing the chassis.  I do not like to remove a chassis if I can keep from it.  Seems like every time I do remove a chassis the faceplate gets bent a little or scratched a little.  Probably because I cannot take my time and be careful enough.  If you want to add a external cab you can with a y connector or just a speaker wire with alligator clips on one end and 1/4" phone plug on the other.

[thelonious]

I've tweaked hundreds of this style Bourns trimmer pot over the last 40 years.

Just for the sake of anyone searching this later (although these links will probably be dead in a year ), here's a similar mil-spec Bourns series (RJR24) that is a quarter of the price. Main functional difference is 3/8" square instead of 1/2" square and 10% tolerance instead of 5%. But we probably don't care too much about either of those characteristics if it fits on the board.

T