what the valves do in a Princeton Reverb II

That said, one of the satisfying things about all-valve amps is their comparative simplicity. Only 5 stages of amplification and the tiny guitar signal is beefed up to potential pain level; and each of those stages is fairly easy to change. In modern solid-state amps, so many chips are used, each one containing hundreds of transistors, the simplicity is lost, the sound must suffer, and experimentation is very difficult.

Read up a little on who makes valves these days. Despite there being many brand names out there, if it's a current-production valve, it came from one of only four factories; one in China, two in Russia, and one in Slovakia. It's embarrassing to get excited about the "differences" between 2 brands and then discover the only difference is in the label. You can pay more for more testing, and that may be a good thing, but don't forget, this only eliminates the ones you didn't buy - it doesn't improve the tone of the ones on sale.

I should probably add that, the more I discuss and read up on this, the more I get the impression that the speaker's more important. If your present valves are working OK, and not old and tired, then changing the speaker will have a far bigger impact on your tone. (Unless you know different; email me...)

No, they're not. They were originally devised as a slightly lower-gain, low-noise 12AX7 or ECC83 (with coiled heaters to reduce hum), but in reality any reasonable-quality ECC83 or 12AX7 will work OK in the three places where 7025 is specified. My PRII is unbelievably low-noise with every 12AX7 I've tried.

(2) Don't touch the glass

You shouldn't touch them when they're hot because you'll be hurt! Duh! But these aren't halogen light bulbs, and the sweat from your fingers won't harm the valve. But don't shock them mechanically while they're hot. Frequent handling can rub the ink off, which might reduce the secondhand value of a famous-name valve. And try to hold power valves by the plastic base when fitting/removing from the amp, so reducing the risk of detaching the base from the glass.

(3) If you use low-gain preamp valves, the power valves have to work harder, which is a good thing

No. If you use low-gain preamp valves, you may or may not get a different tone, which you may or may not like. Then, to get the same audio volume as before, you'll have to turn the master volume knob up to a higher number on the dial. But if it's the same audio volume as before, then the power valves are doing exactly the same amount of work as before. If you want the power stage to work harder without an increase in audio volume, you'll have to use a less sensitive speaker or throw away some of the final output with a power attenuator. BUT if you're looking for tube distortion, a high-gain valve in the first position will probably create clipping early in the signal chain; not always a good idea. A low-gain valve in the first position keeps it clean further into the amp, leaving you with the option of turning up the master volume to overdrive only the power stage. If you're looking for that classic tube distortion you're more likely to get it that way.

V1 on Schematic - Input valve (7025 = ECC83 = 12AX7)
This one is close to the serial number plate. It's used as two separate stages of amplification in one glass bottle. One stage provides the first amplification after your weak little guitar signal arrives, almost exhausted, at the input jack. NB the amp's volume control comes after the first 'half' of this valve, not before, so it's not just a duplicate of your guitar's volume knob. Then comes the bass, mid, and treble controls, and the volume knob; the other 'half' of this valve then provides another stage of amplification. How the tone stack works

You can get different tones, and more or less noise, simply by playing with different combinations of 'volume' and 'master'.

Note that , when there's no jack plug in the front panel socket, the inputs of both halves of V1 are grounded by a switch in the jack socket. This cuts down on noise.

Obviously you want all the valves working well, but the signal passes through this valve twice, while it's still at a low level; so if this one's a bit noisy and tired, the others won't be able to help no matter how great they are. If you're minus a valve cover, don't let this valve go without, as in theory it's the most sensitive to interference.

Oct 2010 - I'm now using a 12AY7 in this position. On an arbitrary scale, 12AX7s are rated '100' for gain - 12AY7 are rated '40'. This cuts down the gain and ultimately the maximum volume you can get out of the amp, but (on my amp at least) it makes the clean sound just a little cleaner. This is not the same as simply turning down the front-end volume control, though I can't explain why! Using a lower gain valve here also cleans up the reverb a little. Anyway, 12AY7s are in current production by (or at least branded as) Electro-Harmonix, Ruby, JJ, TAD, andGroove Tubes. I got mine from Voc Rock in Nottingham, UK - nice guys - they keep some in stock even though they're not always shown on their website. It's not difficult to find NOS 12AY7s either.

This valve is also a dual triode but the 2 little amplifiers are wired in parallel to create one preamp with enough current-capability to function properly. It has 2 jobs.
(1) It amplifies the signal considerably in order to wiggle the springs in the reverb unit. And it's doing it all the time, even when you haven't selected reverb - the 'reverb' knob comes after V3a (below). BUT
(2) irrespective of reverb setting, when you select the 'lead' effect, most of the (massive) signal from V2 is mixed into the sound via V3b, resulting in V3b being distorted (= "preamp distortion) and offering you a kind of valve overdrive sound.

Therefore the reverb depth drops to near-zero when 'lead' is also selected. This is a universal problem on the PRII and I know of no cure. Even if the reverb level could be kept up (maybe with some component value changes?), it would be a reverb of the clean sound only. Ideally the reverb would come after the distortion stage, so you could simulate "an overdriven amp in a big room", but without a redesign this amp will not do that. Buy a reverb pedal.

When in 'clean' mode (non-lead), the PRII has a reputation for useful reverb only from knob position 1-3; turn it up higher and all you get is more mush. If your PRII has the reverb tank mounted vertically, screwed to the side wall of the cabinet, put it in the base of the cab instead. The tank was designed to be horizontal and the reverb quality will improve by this simple operation. But I found by accident that different 12AT7s can affect this, sometimes improving it. Lower-gain valves gives more useful adjustment; I use at 12AU7 in this position for this reason. That reduces the 'lead' sound to a kind of fat boost instead of distortion. The amp will work without this valve - but minus reverb and lead (overdrive). I used to say on this web page that you needn't bother putting an expensive valve here. But I was wrong. In Feb 06; I put an old Mazda 12AT7 in here (with an old Mazda 12AX7 in V3) and the lead sound is much better. Then I put a used Brimar 12AT7 in V2 and the "lead" sound is now very usable. So don't let anyone tell you the "lead" effect is useless; you just have to be lucky with your valves. I buy used valves from eBay, sometimes without even a guarantee that they work. They don't cost much for Mazdas, Brimars and Mullards, and they nearly always work.

V3 - Reverb Pickup and final preamp (7025 = ECC83 = 12AX7)
(Nearest 'pedal red' socket.) Like V1, this works in two 'halves'. The first 'half' picks up the signal from the reverb unit and passes it on via the reverb knob. The other 'half' provides another stage of gain for the main signal path through the amp. This second 'half' is the bit which is purposely pushed into distortion when you select 'lead', so swopping this valve will change the way that effect behaves. (Whether you like this kind of distortion (ie in the pre-amp), or prefer power amp distortion, is up to you; they are different.) I have an old Mazda 12AX7 in here and the lead sound is much better.

After V3 comes the master volume knob, the lead level knob, and the presence knob. NB the line/recording output comes after the output valves, not from here or anywhere in the preamp chain.

There's a link from the secondary of the output transformer. it comes off the leg labelled 'grn' on the schematic and runs back over the top of the schematic, via a 100K resistor, to the inputs of v4. Below the inputs to v4 there's potentiometer (the presence control) with a 0.1uF capacitor going down to ground. The link back from the output transformer is the Negative Feedback Loop or NFB. It feeds a small amount of final amp signal (loudspeaker signal) back into the earlier circuitry, but the feedback is out of phase - it's positive when the signal coming through the amp from v3 is negative, and vice versa. The result is some cancellation - a reduction in gain - and some reduction in distortion, because any errors introduced in v4, 5 and 6 are cancelled out a little.

(Some folks reduce the amount of NFB (eg, increase the 100K resistor) or get rid of it completely (cut the link at some point). The effect is to make the sound 'looser' with more distortion, more noise, more character and the possibility of the thing going unstable and starting to oscillate (hooting with no guitar connected). I have done this and it's fine with simply disconnecting that resistor, but there's not much difference at low volumes, and it disables the presence function altogether.

The presence control is a tone control in the loop. Because it's a negative feedback loop, the tone control has the opposite effect to what it should if it were located in a normal part of the schematic. Turn the presence up = reduce the potentiometer resistance = send more highs in the loop to ground = don't cancel out the normal-signal highs heading for v4 = more highs in the final output, adding a bit of zing that might otherwise get lost.

V4 - Phase Inverter (7025 = ECC83 = 12AX7)
This one splits the signal so that both the signal and its mirror image are sent to the pair of power valves, so they can do their push/pull thang. I read somewhere that gain isn't a critical factor here. However some folks say you want a 'matched' valve/tube as your phase inverter if you want the maximum clean volume. This means the valve is matched with itself - there are 2 separate gain stages in a 12AX7; they are supposed to be identical but of course they usually aren't. Some suppliers will test for good matching of the 2 gain stages and sell the valve as 'matched' (at a higher price). This doesn't seem important to me, especially since the other components are 10% tolerance and not "matched", but I'd be pleased to hear from you if you've investigated it.

V1-V4 do not need to be matched to each other in any way, and when you change them, no internal adjustments are needed. V1,3 and 4 (NOT V2, to be true to the original design*) may be interchanged with each other - the amp will work OK. If these valves come from different manufacturers/batches, the response, tone and noise levels will change as you swop'em around. Turn off the amp and allow to cool before changing them, though.

You can kind-of interchange 12AX7s and 12AT7s. There won't be any damage but the amp will behave very differently. You might like it; you might not. Basically the difference is the 12AX7 (used throughout the main signal path) has a higher gain but lower current capability. The 12AT7 was deliberately specified by the amp designer for the reverb driver because more current and less gain is needed at that point. So theoretically a 12AT7 in V1,3 or 4 will reduce the amp's overall volume capability in return for different tone. As already noted, a 12AX7 in V2 has an effect which some folks like, though I would expect it to die early because it's trying to supply too much current.

V5 and V6 - Output or Power Stage (6V6GTA)
Bigger and fatter than the other valves, and lining up with the back panel where it says 'Princeton Reverb II'. They don't have covers, as they're barely susceptible to interference, and need plenty of ventilation. They work as a team to push and pull current through the output transformer, which in turn moves the loudspeaker. (At the risk of stating the obvious... the amp needs both of these valves to work; if one fails, you lose more than half of the power, with a truly horrible tone.) These 2 valves are best replaced as a matched pair; even then the amp needs checking, and maybe adjusting, as described in Rebiasing the output valves/tubes. Rebiasing isn't simply about nice tone, it's also about keeping your nice new power valves from self-destructing.

These are specified as 6V6GTA. In practice you can use anything beginning 6V6 except those with metal casings. Metal-casing 6V6s have three problems; they're often microphonic (relaying mechanical vibrations as sound out of the speaker), they were designed with lower plate voltages in mind (so are more likely to fail) and they were sometimes constructed so that a high voltage might appear on the casing (so they're unsafe). Use only glass tubes, which what the GT stands for.) CV511 is an alternative name for 6V6GTY. GT, GTA, GTB, GTY... a little work on the web will show you what the difference is. GT's are rated about 15% less power than GTAs. In practice they're all fine so long as (a) you like the sound they make (b) the bias isn't adjusted so hot that they self-destruct.

If the output valves aren't matched to each other, the amp will still work, but in extreme mismatch cases (1) it's not as efficient (2) you might not be able to get a clean sound, and (3) one valve will age faster than the other. But a small mismatch doesn't matter, in my opinion. I have bought unmatched pairs, tested them on a home-made bias probe, found them to be further apart than a matched pair should be, and yet run them with no audible problems at all.

In techno-speak, this amp has a pair of 6V6 valves (tubes) in a class AB output stage with fixed bias. The word 'fixed' here doesn't mean "you can't change anything" - it means you have to take steps to adjust them. The other type of amp design is called cathode-bias; that's self-regulating (bias doesn't need adjustment), but it's not used in the PRII. The combination of class AB and fixed bias gives the most power - in other words, you're probably not going to find a louder 2 x 6V6 amp.

I'm told V5 and V6 can be replaced with 6L6's (which are normally fitted to larger amps). You get more power but at a cost, a 4 ohm speaker is needed instead of 8 ohm, and the amp MUST be rebiased, no exceptions; see the modifications page for details.

Try to hold power valves by the plastic base (not the glass) when fitting/removing from the amp - it reduces the risk of detaching the base from the glass. There is a small 'key' moulded into the centre of the valve to make sure the pins line up with the correct holes in the socket, but some brands of valve can still be misaligned.

Some folks use some kind of clip-on fan to run extra cooling air over the power valves. It can't make any difference to the temperature inside the valves - if it did, they wouldn't work properly! But it does reduce the amount of heat flowing into the rest of the amp. Most of the works live inside an unventilated metal box which is accidentally heated by the power valves, and that can't be good for them, so anything which blows heat away will be A Good Thing. Here's a little project I did... amp-powered built-in fan

Finally...
There's the output transformer. The loudspeaker and the line out/recording socket both come off the output side of the output transformer.

... and that's it. On a really classic all-valve amp you'd have one more valve, i.e. the rectifier, but in this amp rectification is achieved by 4 silicon diodes. There are advantages and disadvantages with each method.

PRII home

How the tone stack works

I am indebted to 'grkeith' on the Fender Discussion Page, who has allowed me to include his description of the PRII tone stack here. It will help to refer to the schematic as you go. He writes...

"There's a lot going on in that area of the circuit. They were obviously really trying to enhance the tone.

Like the common BF and SF Fender tone stack, the Treble control acts as a balance control for the highs and the lows. On the schemo I have, there is an error: the 100k should be shown connected to the plate of the tube and the 250pf. The signal is split there; the highs like to go through the 250pf to the 'upper' leg of the treb pot, the lows are resisted there due to that small 250p cap. The lows however more easily go thru the 100k slope resistor than the highs and divide up between .1 bass cap and .047 mid cap, and get drained off to ground according to the position of the bass and mid pot. The highs don't like to go through resistors as easily as lows, so they are attenuated slightly. If the mid and bass pots are set high, then more of the signal gets put on the 'lower' leg of the treb pot. Thus explaining the standard Fender tone stack and why the treb is a balance pot between highs and lows. So now, all the signal is coming out of the wiper of the treb pot, and sent to the vol pot via a network of caps and res, where more is happening that you don't normally see on a Fender tone stack.

The volume pot is center-tapped. When the volume is at half way, max highs through the 120p and 100p are sent direct to it, less if it's on either side of half way. signal is sent thru the 820k slightly attenuated in the highs again due to the resistor, and goes to the upper leg of the vol pot. The 820k resistor in series with vol pot also loses some volume at the top of the vol pot, and remember some of the highs are already held back in addition cuz of that 820k res.

Now you have another leg going thru the 500p and the 1m, the one meg is shorted out when the push switch on the treb is closed effectively putting the 500p direct to the top of the vol pot, so when the vol is dimed there is a direct line from the treb pot thru the 500p (the 500p holds back the lows from getting thru) to the top of the vol pot.

So there should be an apparent tone shift through the range of the vol pot.

There is a circuit to bypass the 250p treb cap which adds more mids.

That help?

(This message was last edited by grkeith at 06:18 PM, Mar 9th, 2007)"

grkeith, who wrote the above, has his own amp repair business in Grand Rapids, Michigan, USA.
You can see this circuit working (but not the pull-boosts) if you play with the component values in the 'Fender' section of this excellent free downloadable software.

One implication I hadn't thought of is that the entire signal of the amp passes through the treble control. So don't bang THAT knob on a door frame, eh?

Users' Experiences

Here are three ideas. Your experiences may differ. Variations in component values from amp to amp, variations in valve/tube manufacture, variations in bias setting, and variations between your ears and mine all mean this is subjective. I used to detail a lot of users' experiences with various brands, but several years down the line I'd just say, they all sound fine, don't stress too much about getting a NOS for v1 etc etc. Plus, if the amp's working properly but you're really not happy with the sound, tube-swopping won't help nearly as much as a speaker change.

Jan 2012 - ampman91510 on the FDP writes about putting a 12AT7 in v4 (the phase inverter) instead of the usual 12AX7...
"I didn't take a look at the schematic till after the amp arrived, and when I saw the long-tailed PI I went immediately to the stash of 12AT7's. I tried several, and the CV-4024 (that's an alternative number for 12AT7 - Andrew) seemed to sound very nice with this amp. It has a nice breakup, that's one of my most important things I listen for when experimenting with tubes, I like the amp right in the initial mild-breakup area, and if that sounds good then that's probably the tube I'll use. (I play a lot of southern rock, that kind of stuff, sometimes I'll dial up the preamp so it's just barely starting to break up when playing a very loud chord, and then raise the volume till the output stage starts crunching a little, that's just about the sweet spot for me). When the amp arrived it still had the original Fender-branded 7025 in it, and that sounded fine, it was more gainy and maybe a little darker sounding (it sounded nice though, kind of like an old RCA or something). And, it sounded more like a Princeton Reverb with that tube. With the 12AT7 in place, it sounds more Deluxe-ish, it definitely has the Deluxe vibe"
You can try this yourself simply by taking the valve out of v2 and putting it in v4 - you just lose the reverb and 'lead' sound when v2 is empty.

Oct 2010 - I've started using a 12AY7 (lower gain) as V1 and I'm loving the resulting cleaner, lower-gain sound. This is in effect a step back from the 1980s (when there was an assumption that most users would want a lot of gain) to the Fender-Tweedy days of the 1950s (when they assumed the opposite). As a side-effect, the reverb is cleaner too.

I've made myself a variable speaker attenuator ( a very crude imitation of a Weber MASS, or THD Hotplate, or similar). This allows me to use the sound of the power valves running at maximum power, but with the final volume being as quiet as I want. WOO HOO! The folks who write about this are right - it opens up a new world of tone. Just to be clear, this is an extra box which connects between the amp's speaker output jack and the speaker. It dumps actual power - it's not just a 'volume box' which some folks put in their amp's effects loop, which only works at line level. (In any case the PRII hasn't got an effects loop...).

Because a speaker attenuator allows you to run the amp at high power, I wouldn't recommend it if you haven't added screen grid resistors (see the modifications page)

With a PRII and a speaker attenuator, there are 2 basic options.
(1) With the amp set for 'clean' (so you're not selecting the 'lead' sound). I have the volume (input gain) set at 9 or 10. Then turn the attenuator down to low volume and turn the amp master up to 8, 9 or 10. This gives an on-the-edge blues sound.
(2) For a more high-gain, squealing harmonics blues-rock sound, Select 'lead' . Turn the lead level up to 10, turn the attenuator to low volume, and use the volume control to adjust how much preamp gain/distortion you want. Even if you don't like the 'lead' sound, you'll probably like this because it has the 'buzz' removed, and warmth added, by the 6V6s running at near-max power.

If you only ever run the amp on '3' the power valves will probably last forever. A power attenuator allows you to wear out the power valves faster, because you can play at max power all day without hurting your ears; with the master or the lead level on '10' the amp is working to the max whether it sounds loud or not! This makes a clip-on fan a really good idea. It won't save the power valves but it will stop the valves and transformers heating up the rest of the amp so much. Here is my idea for a simple, cheap, quiet, amp-powered built-in fan which doesn't even need you to dismantle the amp.

--------------------

email me... SPAMDEFEATstratopastor@hotmail.com

Trademarks ; All trademarks are acknowledged – Fender, their amp model names, Rivera, Krispy Kreme.

Safety / Damage Disclaimer
Valve (tube) amps develop LETHAL VOLTAGES while running, and store them in charged components EVEN WHILE SWITCHED OFF AND DISCONNECTED FROM MAINS SUPPLY. These voltages are MUCH HIGHER than mains, and higher than anything you’ll find inside a transistorized amp. If this scares you, good. Inside a chassis, don’t use your fingers to touch anything which isn’t insulated or earthed (grounded). Don’t stick more than one hand in at a time, and keep the other hand well away. Use fine-nose pliers to manipulate components. Never, ever work inside a live amp while holding a connected guitar. If you’re not sure what you’re doing, get local help.

The details in this whole site are believed accurate but you act on them at your own risk. I have to disclaim any responsibility for injury, damage, loss of value or loss of gig due to inoperative equipment. Many of the web pages linked from this site say roughly the same thing, and their content is of course not my responsibility.