Help me better understand this 1963 Gibson Invader amp

[waldner]

I have this exceptionally clean example of a 1963 Gibson GA30 RVT Invader.  It's got the original 12" and 10" Jensen's and everything seems to work perfectly but... It's not nearly as loud as I would expect from a pair of 7591s driving two speakers.  My understanding is that 7591 tubes are rated for 19 watts max each, which should make this louder than say a deluxe reverb, but it's not.  It sounds good and very polite with trippy sounding reverb and the deepest best tremolo I've ever heard.

Can anyone take a look at the schematic at the following link and see if they can shed light on a few things:
https://el34world.com/charts/Schematics/files/Gibson/Gibson_ga30rvt.pdf

1. This is the closest schematic I can find anywhere but my amp is different in that it uses a 5AS4 (5U4GB equiv) rectifier and all the schematics show diodes and an OA2 voltage regulator?

2.  The 7591 tubes appear to be grid biased but I don't see a potentiometer to adjust that bias coming off the grids anywhere?

3.  The reverb is weird in that it's like a separate signal that gets added on top of the dry signal.  Sounds ok low but gets out of control pretty quickly.

4. Why is the tremolo so good and deep?

5.  I've seen that people have done a mod to replace what looks like a voltage divider (can be seen once in each channel, has a 500pF bypass cap, and connects to the wiper of each mid pot) with a simple .02 cap.  What are these voltage divider looking things doing exactly in this circuit?

Any insights would be great.  Thanks!

[Latole]

 
 2- Most of the time there are no pot. You have to change resistor (s)  value

[bmccowan]

5. That is what is usually called a T filter. Some call it the Gibson tone sucking circuit. As you saw there is info here, other forums, YouTube as to removing them. Supposedly they were added in an attempt to clean up the sound. I have removed them in smaller Gibsons of that era, and it improved those amps immensely. Earlier Gibsons did not have that T filter.

[bmccowan]

And:
2 - that's right, its fixed, non adjustable bias. Not that unusual. You can change resistors or add a pot.
Volume - if the tubes are good and the B+ voltage is good, and the bias neg voltage is good, the next place I would look is the filter caps. They used pretty crappy caps and placed them in a hot spot. The dual axial F&Ts fit well in there.
4. Tremolo -  They got lucky?

[Latole]

Fixed bias do not mean not ajustable.
It mean the contrary of cathode bias

Fixed bias ; need to be adjust one time and can't move by itself. Cathode bias ( short answer ) "follow" the tube you put in.

In a fixed bias amp, a constant negative voltage is injected to the grids of the output tubes while the cathode is grounded to the chassis. In a cathode-biased amp, the grids are grounded through a grid return resistor and a resistor is placed between the cathodes and ground.

[bmccowan]

Fixed bias do not mean not ajustable.
It mean the contrary of cathode bias

That is correct, but there is fixed bias that can be adjusted, as in the larger Fenders, and fixed bias that cannot be adjusted without changing resistors or adding a pot. Princeton and Princeton Reverbs for example.

[Latole]

Fixed bias do not mean not ajustable.
It mean the contrary of cathode bias

That is correct, but there is fixed bias that can be adjusted, as in the larger Fenders, and fixed bias that cannot be adjusted without changing resistors or adding a pot. Princeton and Princeton Reverbs for example.

You mean there are no pot but it is adjustable. Any amp tech can change resistors to adjust.

[bmccowan]

Yes - "cannot be adjusted without changing resistors or adding a pot"
Fixed bias is an unfortunate term.

[TitaniumValhalla]

I own/have owned a couple of Gibsons with dual 7591's. I have a GA-60 Hercules now and I've had a couple of GA-25RVT Hawks. The GA-25RVT is pretty similar to your amp and nowhere near as loud as I expected it should be, even after some "improvements". On the GA-25RVT, I'm pretty sure the output transformer and speakers were both bottlenecks. The OT was a pitiful little thing. That wasn't necessarily a bad thing, after some mods I got a nice overdrive tone out of the GA-25RVT at a less ear-punishing volume. The GA-60 gets LOUD but it's a 1x15 bass amp.

During the Crestline era (62-65) Gibson was trying all sorts of wacky stuff in their amps and most of it did not sound great. There were a couple of diamonds in the rough but even those need coupling cap changes to open up the ridiculously thin and bright tone Gibson insisted on at the time.

Depending on how much you have in it, might be worth leaving it as-is, but if you're open to mods you could make this thing really sing with a "normal" tone stack and gain stage arrangement, and OT/speaker upgrade.

[Latole]

Y

Fixed bias is an unfortunate term.

No it is because you did no know or study electronics

[bmccowan]

I agree with Titanium about what Gibson did to mess up their amps during that time period. The T filter being one of those. The earlier GA-30 versions are great sounding amps, but quite different. If you are looking to dive in Waldner, I think it is worth bringing these amps up to snuff. I do not think that the changes needed will hurt the value (not high anyway) of the amp. Forum members will help, no doubt. BTW, this amp should have two Jensens - no reason to change those if they are sounding ok. 

[waldner]

In trying to mod the weird "T-Filter" thing, I've drawn up what might be a start to making it more normal.  But in other mods I've seen (those done by Uncle Doug) the T-Filter doesn't attach to a Mid pot like this one does.  By eliminating the voltage divider with the 500pF cap and replacing it with a .01uF cap, I now don't have a grid stopper on the next tube.  Or does the connection to the wiper of the Mid pot act like a grid stopper?  I don't know enough to do this properly.  Can anyone help me draw up this mod so that it might actually work?

[bmccowan]

I don't know that you "need" one, but you might want one. Somebody like Sluckey or PRR will have better thoughts on that. In the meantime, this piece on grid stoppers is good:
https://www.aikenamps.com/index.php/grid-resistors-why-are-they-used

[bmccowan]

BTW - looked in my files - this appears to be close to your schematic, but with a tube rectifier. Might be your circuit.
After Gibson bought Epiphone things got pretty confusing.

[sluckey]

Looking at your modified circuit, I suggest to totally remove the .01 cap and the mid pot. Connect the Loudness wiper directly to the grid of the next tube.

[TitaniumValhalla]

The 1962-64 schematics were all drawn by the same guy with the same handwriting so they're pretty recognizable and in 1965 they changed to the format of what OP posted so these are probably a later model revision.

By the way, Seth Lover said in interviews that he designed all of these amps at Gibson during these years and he drew the schematics for them as well until they hired a "professional" to do it, but I'm not sure what year that was so I don't know which were his.

[astronomicum]

4. Why is the tremolo so good and deep?

There are a lot of variables so a difficult question to answer. The short answer is that your Light Dependent Resistor (LDR) which changes resistance driven by the changing light source from the Incandescent bulb, is providing a sufficient swing in resistance which, if adjusted to be positioned on the right part of the attenuation curve (for this amp / tie-in point), creates good "Depth" (db variation / reduction) in your signal. It appears your circuit is providing a "Good" swing. The key here is that your circuit also appears to be adjusted properly and you are going to want to keep it that way.

I would suggest you take a few measurements of your tremolo circuit while it is working to your satisfaction and keep them for your records. I would do this before making any changes, even installing new tubes:

•   Measure (or have someone measure) the Bias voltage on the 12AU7 (the schematic you posted shows 3.2VDC for that model) with the Tremolo intensity at Zero and Footswitch Open. Take the reading in hundreds of volts. Some recent data I have collected suggests that it may make a difference. If you ever change the 12AU7 that drives this particular Opto-Isolator, you would adjust the bias back to your reference reading to ensure you duplicate its present performance.

This Opto-Isolator uses an Incandescent bulb. As such, it can and will eventually burn out. If you ever change the Opto-Isolator, having the following measurements for reference will help establish a new Bias target which you would then use as long as that particular Opto-Isolator is used.

•   Measure the resistance across the Opto-Isolator LDR with the Intensity at Zero and Footswitch Open.

•   Measure Minimum and Maximum resistance reading across the LDR with the Intensity at Maximum, Speed at Minimum, and again Min and Max readings at Maximum Speed.

Nice amp. Good luck with it.

[AlNewman]

In my very limited understanding, just because a tube can run at a certain wattage, doesn't mean it has to.  How many watts are your tubes running at?  And, if they're running at optimum, how hard is the preamp pushing them?

[waldner]

I've got the amp at my shop so I will be measuring things this weekend.  Thanks for all of the help and getting me some ideas to try.  I will update with what I figure out and maybe some pics of the insides.

In the meantime, here's the back of the amp with the original Jensen 12" ceramic and 10" alnico speakers.  Kind of a cool and unusual array for a stock amp.

[HotBluePlates]

Hopefully this sheds some light on a few things.

...1963 Gibson GA30 RVT Invader.  It's got the original 12" and 10" Jensen's and everything seems to work perfectly but... It's not nearly as loud as I would expect from a pair of 7591s driving two speakers.  My understanding is that 7591 tubes are rated for 19 watts max each, which should make this louder than say a deluxe reverb, but it's not. ...

"Loudness" depends on the electrical power developed in the power amp and delivered to the speaker(s), and the ability of the speakers to turn that electrical power into spoundwaves.

"Dissipation rating" is how much power (as heat) the tube's plate can withstand.  That's different from the electrical power output they can help control/deliver to the speakers.  The amount of power output depends on the voltage & current available from the power supply, and the class of operation of the tubes.

So a "pair of 19w 7591s" isn't sure to deliver "19 + 19 = 38 watts."  They might deliver 10 watts, or 30 watts, or nearly 50 watts (as shown in the 7591 data sheet).



But what about that other aspect of "Loud" - the speaker?  Speakers have varying capability to turn electrical power into sound pressure in the air.  We see this on modern speakers as a Sensitivity rating that says how many decibels of sound pressure are created for 1 watt of power applied to the speaker.

We don't have sensitivity data for those old Jensen speakers.  But the 12" speaker in the photo looks like a C12R to me (I've got one in a 1962 Ampeg Reverberocket and recognize the tiny magnet).  And we can notice a general trend that speakers with small magnets are less sensitive (output less SPL) than speakers with bigger magnets.  For example, look at magnet weight and sensitivity figures for the modern Jensen C12R, C12Q, and C12N.

Notice the sensitivity of the C12R is 92.3dB while the C12N is 97.7dB (for 8Ω).  That extra 5.4dB is huge for loudness.  When using a log scale (as you should for decibels), +5.4dB is "most of the way to +10dB" and +10dB is roughly where we perceive "double-loud."

"Double-power" is +3dB, so +5.4dB is nearly +6dB which is 2x2xPower or 4x Power.  Huh??

     This means 20w applied to the C12N sounds as loud as 80w applied to the C12R.
     Some players intentionally use lower-sensitivity speakers to tame the loudness of their amps.
     Other players use higher-sensitivity speakers to make their small amp bark on stage.

Plug your Gibby into a pair of 100dB Celestion speakers and you won't be complaining of low volume anymore (those speakers are partly why people say AC30s are so damn loud).

[HotBluePlates]

1. This is the closest schematic I can find anywhere but my amp is different in that it uses a 5AS4 (5U4GB equiv) rectifier and all the schematics show diodes and an OA2 voltage regulator?

Welcome to vintage Gibson amps.  They'll produce amps with 1 model name but 8 different circuits, and only issue a schematic for one of the circuits (and that one poorly drawn).  This is just what we deal with when looking at old Gibson amps.

2.  The 7591 tubes appear to be grid biased but I don't see a potentiometer to adjust that bias coming off the grids anywhere?

"Historic Practice" vs "Modern Understanding". "Understanding" is italicized because we usually breeze right past the fundamentals we need to actually understand.

If every tube with a given type number behaved exactly as the data sheet says, they would all be identical.  We would not need a bias adjustment, because we would know that with a given screen voltage and a given grid bias voltage, the tube would idle at a specific operating point & plate current.

So the non-adjustable fixed-bias amps assume that all tubes are "near-enough to data sheet values" and assume the user just plugs in a new tube.

If the tubes installed are not actually "near-enough to data sheet values" then we get less clean output power.  The tubes might overheat when driven, but probably not.  So we really bias to get "maximum performance" in the form of "maximum clean output power."  (This is true whether forum-goers or self-style techs realize it or not)

   -  Early fixed-bias Fender amps did not provide a bias adjustment either.
   -  Later Fender amps did provide an adjustment that move both tubes' bias in the same direction.  It was probably turned while monitoring power output across a load, until the installed tubes delivered their largest clean output power.
   -  Even later Fender amps recognized that the differences between individual tubes was probably a bigger issue, and provided an adjustment that moved the tubes' bias in opposite directions (the "Bias Balance" system).  Again, it was likely adjusted while the tech monitored the clean output power across a load, and set for max clean power (which would like be at/near equal idle current).

3.  The reverb is weird in that it's like a separate signal that gets added on top of the dry signal.  Sounds ok low but gets out of control pretty quickly.

Fender put their Reverb control at the end of the Reverb Circuit, to control how much reverb is added to the Dry signal.

Gibson put their Reverb control at the beginning of the Reverb Circuit.  This is like having Fender's Reverb control full-up, then adjusting how hard the Reverb Circuit is driven by the Dry signal.  As you note, this can get out of hand easily.


Separately, Fender fed their reverb circuit from the signal coming out of their Volume control.  Turn the Volume lower, and drive into the reverb circuit was reduced.

   -  Gibson does not do this; the feed to the Reverb pot comes from before the Loudness control. 
   -  You could turn the Dry signal Loudness off, and still have reverb depending only on your setting of the Reverb pot.

4. Why is the tremolo so good and deep?

"Deep" because it uses an LDR ("optoisolator") which tends to be On-Off in its response.  The schematic symbol implies an incandescent bulb which would smoothly increase/decrease brightness (and so Amp Volume).  However, most LDRs seem to use neon bulbs which fire On-Off and nothing in between.

A second factor in trem depth is it is inserted in the preamp, where signals are small and easily influenced.  Applying trem in power sections is harder due to signal size.

"Good" up to personal preference.

[HotBluePlates]

5.  I've seen that people have done a mod to replace what looks like a voltage divider (can be seen once in each channel, has a 500pF bypass cap, and connects to the wiper of each mid pot) with a simple .02 cap.  What are these voltage divider looking things doing exactly in this circuit?

5. That is what is usually called a T filter. Some call it the Gibson tone sucking circuit. ... Supposedly they were added in an attempt to clean up the sound. ...

"Bridged-T Filter":  R10, R11 and C6 are the "T" while C5 (500pF) is the "Bridge" across the T.  It is simply a circuit that scoops midrange by combining a low-pass filter and high-pass filter.

A low-pass is represented by the Blue dashed line below, while a high-pass is represented by the Green dashed line.  If each filter is set up correctly, they overlap in a way "that scoops the middle."




If you disconnect C6 from ground, the Bridged-T Filter cannot have any effect and you restore the midrange that was scooped. 

But that's like an ∞Ω between C6 and ground.  Instead, you could put a pot between C6 and Ground, and have adjustment over how much midrange is restored.  Which is why they put that 250kΩ pot in there and labeled it "Mid."


Consider that many 1950s amps had no Bass or Treble controls, and that those amps were very mid-heavy (because our pickups are very mid-heavy).  It was "an advancement" to get Bass & Treble controls in the 1960s, and to have that "new mid-scooped sound" in Fender amps.

This Gibby has Bass & Treble controls, but they are not in a tone stack arrangement like Fender used (which is really just a modified Bridged-T Filter, eagled a little to make different bits adjustable).  Gibson used:

    - A Bass control that shaves bass by changing the effective value of a coupling cap.
    - A Treble control that dumps treble  on one end, and boosts gain for trebles at the other end (semi-active?)

So Gibson's plan for the Tone controls doesn't impart that mid-scoop that was popular at the time for amps/music.  So they added one in by employing the Bridged-T Filter.

Some amps implemented the bridged-T filter with an integrated package with multiple legs.  Folks sometimes didn't know what it was, cut it out, and noticed a whole bunch of gain & midrange restored.  Then accused it of "tone-sucking."  Please don't imitate them, since the Bridged-T just "created the tone the designer intended" (even if we disagree with that choice today).


I've built an amp that uses a Birdged-T Filter, but provides 2 controls over its operation:  One is the Mid control to set how deep the scoop is, the other sweeps the center of that scoop to change the amp's voice.  Some of the difference between Fender, Marshall and Vox amps is the center-frequency of their mid-scoop.

[HotBluePlates]

In trying to mod the weird "T-Filter" thing, I've drawn up what might be a start to making it more normal. ... replacing it with a .01uF cap, I now don't have a grid stopper on the next tube. ... Can anyone help me draw up this mod so that it might actually work?

I don't know that you "need" one, but you might want one. ...

Looking at your modified circuit, I suggest to totally remove the .01 cap and the mid pot. Connect the Loudness wiper directly to the grid of the next tube.

There must be something that references the grid to ground, or else the tube will not function.

As Sluckey says, if you're going to remove the Bridged-T filter, go whole-hog and simply connect the Loudness wiper to the next tube's grid.  Now the Loudness pot is the "grid resistor" and ground reference for the next tube stage.

[bmccowan]

Amazing insight from HBP, as usual. I just wanted to add that the prior GA-30RV - no trem and 6V6 - 6eu7s in the preamp and is well regarded. The preamp is interesting in light of this discussion. It's on Doug's schematics page.

[HotBluePlates]

... I just wanted to add that the prior GA-30RV - no trem and 6V6 - 6eu7s in the preamp and is well regarded. The preamp is interesting in light of this discussion. ...

Yeah, that preamp has no mid-scoop and no Bass control.  Just a simple Tone control that can only reduce treble, but not "boost" treble the way Fender did.

"Boost" in quotes because Fender's Tone control acted like a volume pot bright cap when turned towards the top end of its range, which seems to make things brighter by cutting away bass.

I can tell you from having a Supro Spectator amp (whose Tone control also only cuts treble) that amps seem darker with those controls, even if you leave them turned full-up.  Combine this with an awareness that guitars generally need to chop away bass & low mids to be heard on stage & avoid mud... and maybe it explains some of Gibson's choices (that don't sound great by themselves).

[bmccowan]

Bit of a digression, maybe its my Tinnitus but I love those dark amps: early Valcos and Gibsons are some of my favorites. For my amps with full tone stacks, the treble rarely sees anything higher than 10AM and my guitar tone pot is rolled back a bit. Of course, the only stages involved are in my head.

[PRR]

  -  Later Fender amps did provide an adjustment that move both tubes' bias in the same direction.  It was probably turned while monitoring power output across a load, until the installed tubes delivered their largest clean output power....

Over a very wide range, a 'fixed bias' amp's power output hardly changes with bias setting.

What changes is the harshness at LOW volume. Frequently I would run hardly-audible music through and turn up until the harshness cleared or the tubes melted. Had a Fisher which really wanted to run its 12W bottles past 20W to get sweet. Even in 1976 that was unsustainable. The book said 8W at idle but that was harsh and unstable.

A combination of choices by tube designer and amp designer works-out so you can probably just trim for 70% Pdiss and be happy. If there is no trim and soft-play is rare, pick non-adjustable bias so "average production" (with the brand and vintage of tubes you use) runs 55%-75%, median 65%. In the Golden Age all tubes from the same factory (not necessarily same brand; everybody re-branded) would run within 5% most days. The Dark Ages forced a lot of individual bias for smooth sound without redplate, and/or for hum balance.