Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: proaudioguy on February 06, 2024, 09:42:30 am
-
I found this amp in a pawn shop by accident. Got it for a fair price. Has original Fender tubes in V1-V6. Had Jan Phillips 6L6GCs and some form of copper cap, but looks different from the Weber sold currently. I’ll attache photos for that.
I put in a GE 5U4 and a pair of Magnivox labelled RCA 6V6GTs
It has all the original caps but they all measure perfect in circuit, and there are no leaks or bulges. The amp is dead quiet. I had a bias cap on hand so I changed that. I had a cap on hand for V5 Cathode so I changed that. Mine is a larger voltage so its a little large. The one that was in there was 33µF and its 25 on the schematic, so I put a 25µF in there.
I biased the amp. The plate to cathode voltage was 400V with the 6L6s but when I went to the 6V6s it was up to 457V. According to the Rob Rob calculators, I have tube drawing the most at 9.4 Watts. Between 20 and 21mA. That voltage seem a bit high to anyone else? If its that high with a GE5U4, what would it be with a GZ34? Its a 77, and wall voltage was 120V in 77. My wall voltage is 125V.
Amp sounds really good so far. Couldn’t crank it yet, other than what I did in the store on the 6L6 that were in it.
One of the 6L6 bases had the guide pin broken off. I’m guessing that is pretty hard to replace since the tube wires are cut? Anyone ever do that? I have other amps I can use those in but of course I don’t want to have them get inserted the wrong way after I’m gone.
I changed the standby switch position to where it is in all the other Deluxe Reverbs, after the first 2 smoothing caps.
Question, when recapping, should I replace the resisters under the can for good measure? They appear to measure fine. Also, if I do, should I use a 10K where the 2.2K is, which was the BF standard? ANy idea why they dropped that one down to 2.2K?
-
460V on the plates is okay for good 6V6s, but may be a bit high for modern manufacturer types (but JJ6V6S could handle that voltage comfortably). 9.4W is an okay outcome. If it sounds good, enjoy playing it.
-
One of the 6L6 bases had the guide pin broken off.
https://www.tubesandmore.com/products/keyway-replacement-piece-octal-tubes
-
I lifted this chart from a thread over at TAG. I thought it was really interesting.
It shows one DR owner's measurements:
PT secondary voltage; B+ at reservoir; total current draw; and power out at three volume settings. B+ dropped 45V at full power.
-
"Amp sounds really good so far."
Kinda feels like "if it's not broken, don't fix it".
I've tried the Tubes And More things to cope with old GE 6L6GCs that had broken bases. Sometimes they work, sometimes they don't. I ended up having to guide them in by eye.
There's a lot going on here. I think the (Weber?) Cap-Can will deliver higher B+ voltage than a 5U4GB will. And, as noted, line voltage is higher now than it was in 1972.
Finally, going back from 1972 silver-panel era to black-panel era will involve several changes. Proceed accordingly?
-
I would not use that Weber Copper Cap. I've been shocked by one once yrs and yrs ago. Seeing that opening on the side would make me super super nervous. I'd replace it with a tube rectifier.
I am not saying the Weber Copper Caps are dangerous .............. simply reporting my experience.
With respect, Tubenit
-
I contacted weber. They said its not one of theirs. I’m not going to use it until I know what it is. I’ll see if I can get it open at some point.
Let some smoke out while the amp was idle. I had a 47 and 3 22s on hand so I swapped them out while I wait for my 5 16s to arrive, and the cathode caps as well. Its WAY stiffer sounding now. More like the twin. I don’t like that. Amp sounds good but too stiff. I want the solid bottom end, but I don’t want the attack. Not sure how to deal with all that.
Speaker buzzes with low notes. Its not rubbing or anything. The baffle also vibrates a lot. Its one of those glued in baffles. I am seriously considering cutting it out and making a proper 1 piece baffle with grill cloth. I’ll reuse the existing grill cloth if I do that. I’m concerned it will affect the value but I also know it will vastly improve the sound of the amp. I have the exact same issue with my DRII which also has the glued in baffle. I was considering cutting that one out before I bought this one. Its crazy how much bigger that amp is physically than this one.
ANyway other than that this seems like the perfect balance between the twin and the Princeton. The Princeton really does sound like a practice amp now next to this. The low end on this thing is huge. I plugged it into my twin speakers and it has substantially more low end than my twin. I may have reduced the coupling cap size in that twin years ago. I don’t remember.
Anyone reduce the coupling caps in the Deluxe Reverb to reduce the DEPTH of the low end to keep it tighter without adding more filtering on the power supply, which seems to really change the feel for the worst?
I’ve been talking about adding sag to the DRII, changing to a tube rectifier, putting in a resistor, etc, but the sound of this with 47, 22, 22, 22 instead of 32, 16 16 16 is very very much like the DRII with the 80µF disconnected and only the 40, 30, 20, 20 connected. I would not have thought that 4µF per 5 caps (20µF total) would have stiffened up the amp so much. My question for you all, is this. Is it the filtering at the preamp stage or the power amp stage that makes it sound so FAST on ATTACK? I like the softer compressed attack. If I reduce filtering on the V1/V2/Reverb (B+4) and increase it on V7 and V8 (B+1) will I get the tight bass, but also the soft attach?
Can this be tested by placing more filtering in series with the existing? 2 16µF in series is 8µF right? Keep in mind, I am talking about modding the DRII which I have already mostly modded to BF DR territory. The sound is stellar, but its still too TIGHT sounding.
-
Anyone reduce the coupling caps in the Deluxe Reverb to reduce the DEPTH of the low end to keep it tighter without adding more filtering on the power supply, which seems to really change the feel for the worst?
I think the increased attack and "tightness" you're hearing is less related the value of the filter caps and more related to the replacement of the tired originals with new manufacture.
I'm also unsure of your goals. In the quote above you're asking to increase the tightness while in other places you claim to dislike the tightness.
Part of the 'sweet' overdrive sound of the Deluxe Reverb is the big low end in the preamp that causes subsequent stages to fold in on themselves. If you're just going to be playing clean or very little drive, then revoicing to keep things tight is in order. If you want the drive sound, then leave it as is.
One alternative is to replace the cathode bypass cap on the first stage with a 2.2 or 3.3 μF cap. This will gently roll off the very low frequencies, tighten up things near the 100Hz range and keep much of the character of the amp intact.
I normally put this on a switch with a 10-22μF cap in parallel to have both choices available depending on clean vs drive. Works wonders to de-mud a Super Reverb and keep the bigness and chime intact.
-
"The sound is stellar", ...
But "too tight"?
So many variables.
Have you considered trying different speakers, or perhaps different speakers in different cabinets?
FWIW - a few years ago I had an old Jensen 15" frame / magnet assembly re-coned to give me something akin to a Jensen P15Q. The end result was anything but tight, but it was also pretty quiet because of the old, small magnet and the resulting low sensitivity. But loaded with squishy compression.
So many variables.
Ed
-
I'm not familiar with these 'copper cap' devices. I get that they are a type of rectifier, but what is the reason for using one vs. the stock tube rect, or a simple diode replacement?
-
I'm not familiar with these 'copper cap' devices. I get that they are a type of rectifier, but what is the reason for using one vs. the stock tube rect, or a simple diode replacement?
It was my impression that they came about as a "plug & play" alternative to a tube rectifier, mostly for use in older amps.
Octal socket, no soldering, no math...
-
I'm not familiar with these 'copper cap' devices.
It's just a couple diodes and resistor on an octal socket. "Copper cap" name just because of the color of the can. These have been around since the '60s. There are plenty of other brand SS rectifier replacements (https://www.tubesandmore.com/products/vacuum_tubes?filters=646a24c1a646). I have a different brand in my Phoenix amp.
(https://sluckeyamps.com/phoenix/p2.jpg)
-
I see, thanks. I assume copper is used as a heatsink for the dropping resistor?
-
Anyone reduce the coupling caps in the Deluxe Reverb to reduce the DEPTH of the low end to keep it tighter without adding more filtering on the power supply, which seems to really change the feel for the worst?
I think the increased attack and "tightness" you're hearing is less related the value of the filter caps and more related to the replacement of the tired originals with new manufacture.
I'm also unsure of your goals. In the quote above you're asking to increase the tightness while in other places you claim to dislike the tightness.
So with the original caps, the compression and attack was perfect but the low frequencies were out of control making he speaker flop around at high volume. After the caps were changed TEMPORARILY with larger values, essentially 22 22 22 22 22, the BASS is tighter, which is good cause it isn’t flopping the speaker around looking like it will jump out of the gap, but the attack is very strident and the compression “feel” is gone. Feels like the Twin. Its a hard, uncompressed attack. I want the BASS tight and controlled and the rest to be compressed. Not saying I don’t want the bass compressed. I am saying with the old caps the bass was out of control. The damping factor was perhaps very low. SPeaker was flopping around.
"The sound is stellar", ...
But "too tight"?
Ed
The sound was stellar BEFORE I changed the caps but the bass was loose and floppy, but the compression and touch were perfect. The sound is still great, as in frequency response, with the larger cap values, but now the bass is very controlled and tight, and also the compression is much less, the attach is totally different. Its a different amp. Feels totally different. Feels like a mini TWIN. I want it soft. People that prefer GZ34s and larger value filtering caps, would not want what I like. Leo would not like what I like. My concern is that the 16 16 16 16 16 is not going to “feel” any better than the 22 22 22 22 22 that is in there now for 36 more hours. I suspect the old caps being not quite right, is what made it SO good. Now how to I duplicate that without broken caps? Use less filtering for each stage? Then the sound gets better but the bass loses control? How do I get BOTH solid bass and soft compression with plenty of smooth clipping? Some frequencies need to be starving for juice and others need enough to produce lows without making the speaker rattle from over excursion.
-
I see a couple things which "could" have that sort of effect on the compression, besides maybe a speaker change.
If you look at your amp in comparison to an ab763, the grid leaks on the PI are 330k vs 1M. Also the dropping resistor is 2.2k vs 10k. Both of these factors could have an effect on the amount of squishiness at the PI.
-
I see a couple things which "could" have that sort of effect on the compression, besides maybe a speaker change.
If you look at your amp in comparison to an ab763, the grid leaks on the PI are 330k vs 1M. Also the dropping resistor is 2.2k vs 10k. Both of these factors could have an effect on the amount of squishiness at the PI.
Also take a look at the input cap to the PI. Blackface value is 10X lower. This cap forms a high pass with the input impedance of the PI. It's not straightforward to calculate exactly what that impedance is, but it's dependant on the 330k resistors mentioned earlier. On a Blackface these are 1M. So maybe try a .0033μF input cap and leave the PI resistors alone. I also think changing the 2.2k dropping resistor to 10k will get you a softer feel.
I think they were compensating for the voltage drop due to the use of a 5U4GB instead of the 5AR4.
-
I never even noticed the PI input cap. Good catch.
I'd probably still do the grid leaks 1st, and hit the PI a little harder. Then maybe dial in the coupling cap.
-
As far as the attack, I'd be inclined to try a sag resistor off the rectifier. Something relatively small like a 47-68 ohm 10watt would be where I started (and because I have some in stock for AC30 style amps).
I know you're using a tube rectifier but you're wanting to replicate the performance of degraded caps. This might give you that additional softness in the transient.
I stand by my earlier recommendation to lower the cathode bypass cap on the first stage to 2.2 or 3.3 μF.
-
I see a couple things which "could" have that sort of effect on the compression, besides maybe a speaker change.
If you look at your amp in comparison to an ab763, the grid leaks on the PI are 330k vs 1M. Also the dropping resistor is 2.2k vs 10k. Both of these factors could have an effect on the amount of squishiness at the PI.
I see a couple things which "could" have that sort of effect on the compression, besides maybe a speaker change.If you look at your amp in comparison to an ab763, the grid leaks on the PI are 330k vs 1M. Also the dropping resistor is 2.2k vs 10k. Both of these factors could have an effect on the amount of squishiness at the PI.
Also take a look at the input cap to the PI. Blackface value is 10X lower. This cap forms a high pass with the input impedance of the PI. It's not straightforward to calculate exactly what that impedance is, but it's dependant on the 330k resistors mentioned earlier. On a Blackface these are 1M. So maybe try a .0033μF input cap and leave the PI resistors alone. I also think changing the 2.2k dropping resistor to 10k will get you a softer feel. I think they were compensating for the voltage drop due to the use of a 5U4GB instead of the 5AR4.
YES! I didn’t change the PI but I did change the dropping resisters and used 10K for both when I recapped it with 16µF tonight. WOW. its great. the bass is tight and the sound is compressed and squishy. The amp has a substantial amoutn of bass before and that is decreased to something managable. I also found all the speaker screws loose on the FRONT of the baffle. The nuts on the back were tight except for 1. I cranked the screws in the front and that seems to have helped a lot with the buss/ rattle along with the speaker not flopping around now that the bass is controlled. I wonder if the cathode bypass caps being out of spec was affecting the bass resonse because the 22µF caps I had in there the last few days were F&Ts. I can say with better clarity, if you want a more TWIN like sound from the DR, use a GZ34 and 22µF filter caps, and if you want more compression use a 5U4, 10K dropping resisters, and 16µF filters.
I would love to have more detail on those 330K grid leaks and how they work vs 1M. What affect does that have on the response? I’m going to hit the ROB ROB site and see what it says on that but feel free to fill in the blanks. I am taking everything I’ve learned here and applying it to the DRII.
-
You need large changes in value of a cathode bypass cap to really hear a difference. Depending on the cathode resistor and bias point of the triode anything larger than about 8.2μF will be hard to distinguish from each other for guitar frequencies in all but the most controlled of settings. If your cap measured 30μF or 15μF when labeled 22μF, you would not hear a difference.
The PI question is harder to answer since the Long Tail Pair is a complex circuit. The easiest way to think about it from a frequency perspective is the input impedance with 330k grid leaks is approximately a third of that with 1M grid leaks. So to keep the frequency response the same as a Blackface you need to scale the input capacitor accordingly.
To calculate the actual cutoff would require you to understand the bootstrapping effect of the circuit. There are good sources on this. The Aiken white paper discusses this is detail.
Some designs use both values 330k grid leak on input 1 and 1M on input 2. I think the goal here is to balance out the low frequency cutoff points between the two inputs to the PI. Since most of these cutoffs are below the lowest guitar fundamental (except maybe the ones with a 500pF input cap on some Fenders) I think a lot of this is moot for the same reason that applies to the cathode cap discussion above.
-
If your cap measured 30μF or 15μF when labeled 22μF, you would not hear a difference.
How about a 330uf bypass cap? any difference ?
-
Not in the guitar range. Maybe on the lowest of low notes on a 5 string bass--assuming your speakers can even output that.
Those caps in the 100s of μF tend to be there for reasons other than frequency response. They can reduce noise in some circuits for reasons I don't quite yet understand. Others will be better sources of info here.
-
I think the most common reason for using 250μF-330μF bypass caps for v1 cathode is to reduce heater-cathode leakage noise.
/Max
-
after much experimenting, anything over 47uF is showing off. anything over 47uF is hiding a problem you didn't want to fix
-
If your cap measured 30μF or 15μF when labeled 22μF, you would not hear a difference.
How about a 330uf bypass cap? any difference ?
F=1/(2*pi*R*C)
-
F=1/(2*pi*R*C)
The above equation needs a bit of context. That's the general equation for calculating the corner frequency of a passive filter network.
It's also known as the -3dB point because that is the frequency at which the source signal is cut by -3dB. However, different from a traditional high or low pass network, partially bypassing acts more like a shelving eq than a constant roll off by XdB/decade. The degree of the shelf is about 6dB, so the calculation gives you the midpoint and no real info on the slope.
This is important to know because a partially bypassed triode at say 100Hz will affect the signal well into the 300Hz range, but won't completely kill the lowest of frequencies. It'll just roll them down. This is a different effect than just shrinking the coupling cap.
With an 8.2μF cap and a cathode resistor in the usual range of 820-2.7k the -3dB point is 2 octaves or more below the low E (~82Hz). This is why anything larger will have no impact on the guitar frequency range--it is already fully bypassed for the expected signal bandwidth.
With smaller bypass caps than 8.2uF you can start attenuating the lows, even if the filter frequency is below the lowest fundamental.
This is where things get fun. The first thing a studio engineer or decent sound man is going to do to your mic'd amp is put a high pass on it, and probably way higher than you'd want - 150+Hz. The reason for this is to keep the guitar out of the bass and drum frequency range.
You can do this yourself my tailoring the cathode cap of your first stage. I like 2.2-4.7μF for Blackface Fenders and their derivatives depending on speaker arrangement. 1 speaker gets 4.7μF where a 4 speaker Super gets 2.2μF.
This gives you a better mix when playing in a band unmic'd or with a bad/lazy sound man.
There's also the school of thought that operates under the observation that lowest frequencies require the most power to reproduce and by cutting the lowest frequencies early in the circuit you free the amp up to focus more power into reproducing the useful frequencies.
I just like that the amp sounds less muddy with a gentle roll off of the subbass
-
If your cap measured 30μF or 15μF when labeled 22μF, you would not hear a difference.
How about a 330uf bypass cap? any difference ?
F=1/(2*pi*R*C)
I ask because Marshall bypassed the first amplification half in the 1959 JCM800 Super Lead and others. Probably why I have some that have been sitting here but never used them.
https://el34world.com/charts/Schematics/files/Marshall/Marshall_1959_jcm800_super_lead_pre_amp_100watt_year_1988.gif (https://el34world.com/charts/Schematics/files/Marshall/Marshall_1959_jcm800_super_lead_pre_amp_100watt_year_1988.gif)
-
Pretty sure that's a Bassman carryover.
The 1959 and JCM800s are very different preamps. You'll find no 330s in the 800s
Edited. My mistake
Got confused with the 1987 and 2203/4s
-
Would it be alright to combine a 22 with a 16uF in the reservoir position?
-
Well, from what I've read, basically the 10k vs 2.2k drops your plate voltage at your PI (and pre amp) fairly substantially. Meaning your Pi will start to break up sooner.
The 1M vs 330k would pass more signal to the grid of the PI, where again, it would start to break up sooner. I've done this exact mod to my silver face super reverb with good results, and I've made it switchable. It may actually seem "tighter" with the 1M grid leaks, but that could be subjective. There is a definite difference in gain and distortion between the 2, although it isn't night and day. If you like what your amp is doing now, leave it and enjoy it.
-
Would it be alright to combine a 22 with a 16uF in the reservoir position?
You mean in parallel for a nominal 38μF?
There's no reason not to.
-
Would it be alright to combine a 22 with a 16uF in the reservoir position?
You mean in parallel for a nominal 38μF?
There's no reason not to.
Caps in parallel add values, but the working voltage will always be equal to the lowest voltage rated part.
Caps in series have the same "reciprocal of the sum of the reciprocals" relationship as resistors in parallel. RoT is that the final aggregate value will be smaller than the smallest value in the group. But voltage sharing can be unpredictable..
-
Thanks for information
-
You might play around with this calculator to get a graphic representation of the effect of different by-pass capacitor values.
https://www.ampbooks.com/mobile/amplifier-calculators/cathode-capacitor/calculator/ (https://www.ampbooks.com/mobile/amplifier-calculators/cathode-capacitor/calculator/)