Fender Tweed Twin Reissue JB Edition

[pdf64]

No matter how many power tubes are put in there, HT current can't exceed the limit imposed by the HT voltage and the load impedance.
Tube guitar amps are designed as a system, the power supply, power amp and power tubes all work hand in glove; that's why they sag and bloom when pushed, rather than immediately enter hard clipping.
A TR AB763 has a higher, much stiffer HT than a 5F8A, a lower load impedance (my guess), and beefier power tubes.
Of course a GZ34 would likely be overstressed if one was used as the rectifier in a TR AB763.
But a TR AB763 is clearly a different design to a 5F8A, the fact they both happen to have 4 power tubes is immaterial, eg Vox AC30 and Marshall Major also have 4 power tubes.
Why not trust that the designer of the 5F8A was competent and hence the parts specified are fit for purpose?

Please note that the 'bias to idle at 70%' guideline would seem to be misapplied when tubes that have a higher rating than those that the amp was originally designed around are fitted.
Hence when fitting 6L6GC to amps design around earlier, lower rated 6L6 types, for the amp to work as the designer intended, it would seem to be wrong to bias at a hotter operating point than would be appropriate for the intended tube type.

[Roe]

the OT has a 2k reflected primary impedance on a 5f8a, just below the 2k1 or so on the ab763 twin reverbs.
250ma is minimal for the 5f8a circuit, but fender was limited by the tube rectifiers. they even experimented with mercury vapor or somthing pretty dangerous before they gave it up. Today, a marshall style power supply seems like a good alternative if you avoid high voltages. the jmp 100s had a 290ma PT and used a SS rectifier. the mains voltage sag 20% or even more when you crank the amp

[pompeiisneaks]

Depends on the tube, but according to this Svetlana tube datasheet, https://www.newsensor.com/pdf/svetlana/ ... etlana.pdf max signal voltage for a tube should be 66mA 66 X 4 = 264mA but that's in Class A operation, of course that's not even looking at preamp t ubes (although they don't pull close to that amount of current). and that's at max power diss. I'd think it's likely going to run much closer to 50mA when really pushed so you've got 50mA to spare in most use cases. If it's pushed at max volume with square wave input it may be unhappy.

~Phil

[pdf64]

The HT winding voltage of PTs intended for tube rectifiers sags a lot under load.
They have to, rectifier tube info specifies a minimum equivalent plate resistance, and that resistance is provided by the HT winding.
See the chart at the bottom of p4 http://www.mif.pg.gda.pl/homepages/fran ... 5/5AR4.pdf
For this amp, curve 3 seems the closest and so we should expect ~105 ohms per plate equivalent resistance in the PT HT winding.
Hence such PTs have poor regulation / are necessarily saggy and will run hot (the HT winding alone will dissipate at least 25 watts under load).

And so the 400Vdc HT at idle will likely drop at least 10% at full unclipped output and >20% at max square wave, especially when considering that the increased HT ripple at such loads will bring the average Vdc down even further.

As VHT drops, the max iHT that can be pulled through the OT primary impedance will necessarily also fall.

http://schems.com/schematicheaven.net/m ... _schem.pdf

[a j shaner]

Thanks everyone for all the information.

Perhaps I have not been clear about my concerns regarding the 5AR4/GZ34 rectifier.
If I am correct, The 5f8 and 5f8a Twin Amp was Leo Fender's first attempt at making a
4-6L6 amp. The urban legend has it that this is an 80 watt amp.
The 5f6 and 5f6a Bassman Amps were rated at 40 watts.
While I have not had access to a 5f8a, I have built two 5f6a amps and they are much closer to 26 watts RMS under load.
When the first reissues of the Bassman came out they were solid state rectified with a power transformer rated at 375-0375@200mA.
I built one of these and understood the complaint of a lot of players-They were overly bright and brittle sounding.
Even the one that I built with a 5AR4/GZ34 rectifier had very high plate voltages(app.490 on the plates). This is far from the original spec of 425.
The one that I have now is almost dead on spec and rates about 26-28 watts RMS output.

Now to the question of the 5f8a twin.
The twin is an odd bird, there are not many of them around. It seems to have been manufactured as an afterthought.
According to a number of amp pundits the twin amp had a tendency to blow rectifier tubes when pushed hard, whether this is factual is hard to confirm.
It seems that the Twin Amp was overlooked in the transition from Tweed to Brownface, 1959-60 saw the development of a new style of Fender Amp, starting with the 5G13 Vibrasonic.
The use of solid-state rectifiers became common practice, especially in higher power amps. This makes sense considering that Fender was marketing to Southern California country musicians.
Rock music had not yet taken hold the way it would after the British Invasion.

There is(or was) a website called the 5G8 Twin devoted to the mystery 40 watt small box Twin Amp(It appears to be a Vibrasonic with 2-12's and a vertically slanted Twin Amp logo).
One would think this was a prototype that never saw production as the 6G8 and 6G8a Twin Amp followed in 1960 and never had the reverse control panel layout.

Now that I've digressed into Fender lore, the point was that tube rectifiers were insufficient to provide the necessary current for high power amplifiers. The physical space limitations of Fender Tweed Amps did not allow the use
of dual rectifier tubes. Also production costs would favor the use of solid state rectifiers.

Going back to my original post, The use of a 5V3a/5AU4 rectifier appears to be a viable alternative in the 5f8a Twin Amp provided there is sufficient 5V filament current to run the rectifier.
Using the RCA RC-21 tube manual the maximum design value for the 5V3a is 415mA under load(470 unloaded) in conjunction with a quartet of 6L6GC's (230mA current zero signal, 420mA max signal, from RCA)
this would seem to possible for use in a one-off personal project.

Martin Manning, if you read this, your opinion on this would be greatly appreciated.

Thank you everyone, it's all good fun.

AJ

[a j shaner]

https://courses.physics.illinois.edu/ph ... n_Amp.html
The 5G8 Twin Amp site

[martin manning]

5F8-A schematic shows plate voltage at 400V and 4x 5881. I don't know what the plate load is, but 2k a-a is likely. That's about 100W flat-out, so 80W seems reasonable for clean power output. If you use 6L6 instead of 5881, the output power (and current demand) will be about he same if you run the same voltage and load. I think you have an idea there using a 5AU4: 325mA at 400V vs. 5AR4's 250mA at 400V. It does requires 3.75A filament current vs. 3.0A for 5AR4, which would be a stretch for a PT with a 3A rated 5V winding. However, Mojotone's 5F8-A Export PT (MOJO753EX) spec sheet says that its 5V winding has 6A capacity (and 325V at 250mA on the HT). Unfortunately the spec sheet for their standard 120V version doesn't show any current ratings, but you could ask (I often find them doing a poor job of providing specs on the stuff they sell). The voltage drop will be higher for the 5AU4, though, maybe 15-20V(?).

[Roe]

Plate load is 2k. With low voltages this wont produce much power. Most report from 60 to 80 w

[pdf64]

My understanding is that the convention is for the typical Vdc noted on the schematic being under idle conditions; due to the factors previously noted, as signal level rises, the HT will drop significantly, likely to be under 350Vdc at max square wave.

The most likely cause of catastrophic failure of the rectifier tube is that the standby 'hot switches' the reservoir caps, almost certainly exceeding the rectifier tube's peak current limit; don't use standby.

[martin manning]

Plate load is 2k. With low voltages this wont produce much power. Most report from 60 to 80 w

Sorry, typo on my part above, now fixed. Estimated output power was/is correct for 2k load.

The most likely cause of catastrophic failure of the rectifier tube is that the standby 'hot switches' the reservoir caps, almost certainly exceeding the rectifier tube's peak current limit; don't use standby.

Easy enough to move the standby switch to the other side of the reservoir. Fender didn't seem to catch on to that until after the change to solid state rectifiers. Just avoid switching it hot.

[a j shaner]

Thanks to all for the feedback.
The position of the standby switch makes good sense.
60 watt output seems more reasonable.

Once again, thank you for the information.

AJ

[John_G]

Hi,
The series

Twin-Amp 5F8A.jpg

diodes before the tube rectifier should help limit inrush current.
John

[martin manning]

The series diodes before the tube rectifier should help limit inrush current.

The series diodes won't provide any current limiting function, but they will back up the vacuum rectifier if it fails shorted and prevent collateral damage.

[John_G]

Martin, Thanks for correcting me.
I install these diodes in just about every modern amp that uses a GZ34 rectifier. The most common being the Vox AC30 CC where both the Sovtek and JJ tubes blow often. Since installing the additional SS diodes the GZ34s longevity has been greatly extended.
Maybe we are talking more about the improved PIV by using the additional diodes thus protection the tube ??????
John

[martin manning]

Indeed increased PIV is another benefit of the added Si diodes, and could be the reason you are seeing longer life from vacuum rectifiers.