Fender 5E8-A choke mod

[Tweedguy]

I have a question regarding whether or not a choke move is a good thing for my Twin.

I built a 5E8-A Low Power Tweed Twin. I read that if I moved the choke and one filter cap to emulate the High Power Twin/Bassman F series designs, I'd get a better response from the amps power section. The idea is, the screen current is higher than the plate current on my stock design, switching it around would balance things out. Also, the fact that all current from the amp is going through this choke was supposed to be a bad thing.
So I did the mod and also added the 470 screen resistors just like the F series. Originally the screen was about two or three volts higher than the plate, now I've got 445V on both plates and screens.

Here's the issue; I was just told that the original choke placement is necessary because my amp has a Split-Load phase inverter. He said it had to do with current and the inefficiency of this phase inverter. I didn't get it.
I need to know if the stock way is necessary to this design and if not, the pros and cons of each way.

Thanks to anyone with an answer

[Ears]

I'm not going to get into the subject of the choke placement but rather point out that some of your observations appear confused.
For normal tube operation screen voltages shouldn't exceed plate voltages and anode current is usually (much) higher than screen current even at idle in class AB.

[Tweedguy]

I made a mistake when I said the screen current was higher, I meant to say the applied voltage to the screen was higher. It's like this on all E series Fender designs with the choke in this position. I've heard it's supposed to be bad because it inhibits some electrons from reaching the plate inside the tube.

The question remains, is this type of choke placement NESSESSARY to amps with a split-load phase inverter?

[Mikka]

The question remains, is this type of choke placement NESSESSARY to amps with a split-load phase inverter?

No, not necessary !
And I think that it's better to have a screen voltage a few lower than the plate voltage, so you 'll blow your tubes less quickly. In the same time, if you pick up a few higher voltage for your plate, it will increase the power and the headroom of your PA. I tooke a look on the 5E8A schematic and I saw no risk for that.

[Ears]

My 2 cents: In the 5e8-a schematic DC losses in primary windings combined with absence of screen resistor will mean screen voltage is slightly higher than than plates. The p-p amp is far more sensitive to noise in the screen supply than in the plate supply so there is no significant advantage to smoothing the plate supply as in this arrangement. Also anode current in pentodes and beam tubes is largely independent of anode voltage in the normal range of operation (~ horizontal plate curves). The chosen screen voltage has great affect upon power output but I think it depends on the tube type as to how high it can go to deliver real advantage; for example, in big bottle KT88 the data states that there is NO advantage in screen voltages over about 300V.

Running screen voltages higher than anode risks the screen usurpng the role of the plate and also of the screen collecting secondary emission from plate , so drawing excess current.

As always I'm in here to learn so, if somebody disputes the above please jump in and correct me.

So yes, I'd change the arrangement, didn't Fender ultimately do so anyway?

EDIT: But if you change choke position you might want to also up the reservoir cap to better smooth the anode supply, keep it within rectifier specs.

[LOUDthud]

My first question is what are the specs on your choke? Most chokes sold for Fender amps have too much resistance to be used before the OT connection. You would really like something like 2Hy, 50 ohms, 300mA. For a screen choke, 5 to 10Hy, 250 ohms, 50mA would be a better choice.

The more common screen choke does a marvelous job of getting rid of the 120Hz at the screen node. The Ripple is so small, you can't see it with a scope. Under overdrive conditions, the screen voltage will bob up and down but won't have ripple on it like the main B+.

When the plate voltage of a power tube drops to the 50 to 100 volt range due to signal conditions, screen current rises exponentially. It's the job of screen resistors to limit this to safe levels. If at idle, the screens are 5 volts higher than the plates, it doesn't really matter much IMHO.

An amp with a Pi filtered main B+ will sound different than the traditional screen only connection. There is a certain amount of 120Hz that creeps into the overdrive tone that many players have come to expect. When it is absent, they sense that something is missing. Neal Young is one guy that noticed the difference between 60Hz power in the US and 50Hz when on tour in Europe.

I don't think the Pi filtering had much to do with the split load phase inverter but I could be wrong. You need a fairly high B+ on a split load PI to drive 6L6's and poor regulation of the B+ is going to affect one 6L6 harder than the other. So it might have been one of those last minute changes to remedy a problem that later became unnecessary.

[Ears]

When the plate voltage of a power tube drops to the 50 to 100 volt range due to signal conditions, screen current rises exponentially. It's the job of screen resistors to limit this to safe levels. If at idle, the screens are 5 volts higher than the plates, it doesn't really matter much IMHO.

Thanks Mr Thud for this good post. When I read descriptions as above sentence I get a little confused as to whether you are refering to low instantaneous plate voltages or to sagging DC plate supply under large signal conditions. As I've never measured it, does the supply atually sag to such low levels? That would be very poorly regulated.

RCA manual has hifi cct with screens 5V higher than plates as in Fender's but states that the supply has to be well regulated, in part due to fixed bias, which is achieved with a choke before B+ as in 5e8-a. But the RCA example includes series screen resistors.

Also on subject of RCA recommendations, they state series screen resistors shouldn't be used to supply screen potential under high output conditions because of distortion and change in output power. But I guess that outcome (dist. and compression) actually exactly what many guitar amp designs seek to achieve. It well illustrates many of the conflicting criteria that must be juggled by designers.

[Andy Le Blanc]

the statement about the inverter deosnt seem right
a split load has better balance than the paraphase inverters
they replace in the early fender models but the trade off
for this (cleaner tone) is less gain by design not by power supplied
a tube amp can be very simple..... its the application you design for
that dictates the shapes of things..... I've heard and read that Leo was
very observent of design trends and followed courses seen at trade shows
as well as listening to the responces of promotional bands
so rapid changes in fender designs towards a better sounding
and more reliable product seems a better explantion of why the design
was tried... a tube amp is a power supply attached to a speaker
I like to see an amp with these power supply hookups A/B through a
spectrum anyliser

[redtoploader]

My first question is what are the specs on your choke? Most chokes sold for Fender amps have too much resistance to be used before the OT connection. You would really like something like 2Hy, 50 ohms, 300mA. For a screen choke, 5 to 10Hy, 250 ohms, 50mA would be a better choice.

FWIW- the choke Fender used on 5e* amps was a 5-10hy(swinging I was told), 125ohm, 200mA choke. Triad part #14684.

[Andy Le Blanc]

not sure about that.... its a capacitor input filter stage
youd have to check the math to chose the dc resistance
and henries for max attenuation at 120hz

[Andy Le Blanc]

did some digging......
the best reason I can find to justify feeding the primary with a power tap
after the choke is if the primary winding on the OPT has a really low inductance and cant buck the supply ripple......

[LOUDthud]

Triad has a lot of old catalogs online but all the chokes start with a C so I don't think 14684 is a Triad number unless there was a special number for OEM's. The closest thing I can find is the C16A, 10Hy 200mA 150 ohms. It's a pretty big part. Roughly a 3 inch cube with 4 mounting holes that weights 4 pounds. Does this sound right?

All the swinging chokes are rated 5/25 or 4/20 Henrys. They are between 4.5 and 17 pounds.

Ears ... Yes I was speaking on instantaneous plate voltage. Back in the 50's nobody ever imagined amps would be driven into clipping the way of AC-DC or ZZ Top. The spike in screen current wasn't seen as a problem, especially in HiFi amps. So you have to take things in context when reading old texts. Early Fender designs didn't have them for the most part.

[jaysg]

I used a Hammond 159S

4H, 225mA, 65 ohms, 500V.

[redtoploader]

Triad has a lot of old catalogs online but all the chokes start with a C so I don't think 14684 is a Triad number unless there was a special number for OEM's. The closest thing I can find is the C16A, 10Hy 200mA 150 ohms. It's a pretty big part. Roughly a 3 inch cube with 4 mounting holes that weights 4 pounds. Does this sound right?
.

FWIW, if you are referring to the archive that Triad has online- i've never seen any part#'s for any of the transformers Fender used in there. I'm assuming that there are OEM#'s like you suggested, and the transformers sold to the public off the shelf from the catalogs are different. I was hoping to find these in those catalogs and see specs, especially the OTs.

Fender used many Triads in the tweeds and the very early blondes/browns- 7993, 8087, 8160, 45216, 50246, etc...and they all have black cases or frames unlike the grey ones you see in the catalogs. Go figure.

EDIT: heres one that close on ebay:http://cgi.ebay.com/One-TRIAD-14527-8H- ... dZViewItem

The 14684's i've seen in the flesh were open frame. All of them about the size of a 2x6l6 sized OT.

[LOUDthud]

Somebody on another forum had this picture of a 5E7. It looks like the closest Triad common part is the C14X which is still available, now made by Parallax. Back in the 50's Triad had at least two factories in the LA area so Leo would have no trouble getting the parts he needed. Now I think they are in Indiana. I think Weber VST has a knockoff of the 14684.

I think the 150 ohm resistance gives the amp too much sag for todays players. But you can go down quite a bit on inductance (and resistance)and make up the difference with higher capacitance.