1/2 power switch data

BobD · Post by **BobD** » Fri 04/21/06 5:05 pm

Since there has been some discussion about 1/2 power switches, I thought I would share a few measurements from my 36 watter build. The switch is the "Matchless style" in that I use a 62 ohm resistor for full power and add a 68 ohm in series for a total of 130 ohms in 1/2 power.

In full power mode I measure 324 vdc on the plates of the power tubes and 10.1 vdc across the 62 ohm cathode resistor. This calculates out to a plate dissipation of 12.8 watts but since I neglected to remove the contribution of the screen dissipation, I'm ok with that.

In half power mode the plate voltage jumps to 340 vdc and I measure 11.2 vdc across the 130 ohm resistance. This comes out to a toasty 14.2 watt plate dissipation. I won't be using the half power switch until I can get an 82 ohm resistor to swap with the 68 ohm.

I really like the 1/2 power switch, the only real concern that I have is the 340 volts on the plates. That number should come down somewhat once I get the bias cooled down but I'm not sure how much. I'll let you know once I swap it out.

Hope it helps.

markd · Post by **markd** » Fri 04/21/06 8:13 pm

BobD wrote:
In half power mode the plate voltage jumps to 340 vdc and I measure 11.2 vdc across the 130 ohm resistance. This comes out to a toasty 14.2 watt plate dissipation. I won't be using the half power switch until I can get an 82 ohm resistor to swap with the 68 ohm.
I really like the 1/2 power switch, the only real concern that I have is the 340 volts on the plates. That number should come down somewhat once I get the bias cooled down but I'm not sure how much. I'll let you know once I swap it out.
Hope it helps.

I wouldn't expect the plate voltage, at idle, to come down if you increase the cathode resistor value. That will just bias the amp colder and draw less current through the power supply. Less current means less sag in the PT and tube rectifier, so you will get higher plate voltage not lower.
So most of your plate voltage increase happens in the power supply not the power amp section. If you had two EZ81's (or 5V4's), and one could be switched out (or even just pulled out) with the half power switch engaged, you could get more correct voltages.
Or you could just accept the 14 watts as being typical for many original 18watt amps and not worry about it.
mdurham

Chubsman · Post by **Chubsman** » Fri 04/21/06 8:18 pm

How do I calculate that with my amp...?

Take the Plate DCV and Cathode resitor DCV and Value...and do what???

BobD · Post by **BobD** » Fri 04/21/06 9:24 pm

Yup, my plate voltage will go up slightly, I need to think more before I type

As far as accepting the increase in voltage, there's no choice but to accept it if I want to run at 18 watts. I just wonder what that increase in voltage will mean to tube life-it's getting pretty high...

What do you mean "So most of your plate voltage increase happens in the power supply not the power amp section" ? The measured increase was at the plates of the el84 tunes. This is the power supply to the power section

Chubsman, what do you want to calculate - the bias of your power tubes ?

Chubsman · Post by **Chubsman** » Fri 04/21/06 9:49 pm

I too have a 36w with 1/2 power switch... but I am affraid to use it, as Blaren is too....

If I can calculate what you are talking about, I can make it safe for the power section....

markd · Post by **markd** » Fri 04/21/06 10:40 pm

BobD wrote: As far as accepting the increase in voltage, there's no choice but to accept it if I want to run at 18 watts. I just wonder what that increase in voltage will mean to tube life-it's getting pretty high...

340v on the plates and 11.2v on the cathodes sounds good to me. I have measured several original Marshall 18's. With the wall voltage adjusted until the heater windings are at 6.3v (to make sure the voltages would be correct) those amps had very similar readings. Including the 14.2 watts of dissipation. This is part of the 18watt tone, IMO, and tube life will suffer a bit but........

BobD wrote: What do you mean "So most of your plate voltage increase happens in the power supply not the power amp section" ? The measured increase was at the plates of the el84 tunes. This is the power supply to the power section

When you have 4 power tubes drawing their idle current through the power supply(full power) there is almost 90mA of additional current being drawn through the power supply than when you only have two power tubes drawing current (1/2 power). A PT has resistance in the windings and that additional current will drop some more voltage across the PT (PT's have "sag" even if there is no tube rectifier).
I believe your amp has a 5AR4 tube rectifier. That 90mA of additional current at full power will drop most of the voltage difference (340v to 324v) across the tube rectifier. That's where most of the sag in the power supply occurs.
So a higher B+ out of the power supply, in 1/2 power mode, will give you higher voltages throughout the amp including the plates of the EL84's.
Switching between 4xEL84's and 2xEL84's means a compromise must be struck in the power supply, not just in the power amp cathodes. A standard 18watt power supply can't supply the current for 4xEL84's. And a 36watt power supply is too stiff to supply the correct voltages for an 18watt. Switching dual rectifiers is a complicated answer when simple acceptance of the increased dissipation is so much easier. ha! I'm sure the tone doesn't suffer.

I consider the use of a choke in an 18watt style amp, without the 1K shared screen resistor, to be harder on EL84's than using a 36watt at 1/2 power.
Now, if your using the half power switch and you are using a choke without the shared 1K screen resistor that's just asking for it IMO.
mdurham

Post by **Plexi** » Fri 04/21/06 11:01 pm

Mark.. that might work pretty good.. if you wanted to run the amp 1/2 power and not kill tubes.. also use a 5V4 instead of the GZ34..
this just might get things to be where they should. The voltages Blaren posted ,and also the amp i had here..was more like 354plate and 348v screens in half power..
The livepool half power switch works pretty good..but that amp runs much lower voltages,and easier to deal with. Only disatvantage on the 36 would be..you'd have to decide before playing the amp..or let it cool off to change the recto tube. I just picked up 5 NOS RCA 5v4 for 15.00 the 5U4 crrent is a bit much on the edge for the 36 PT,plus it has more sag.
But anyway i like your idea about changing the recto tube to run in 18 mode.

Richie

madkat · Post by **madkat** » Sat 04/22/06 1:35 pm

I like the idea behind the 18/36 watt switchable option (maybe my son will go deaf that much slower) but what is the effect on output impedance with 2 vs. 4 output tubes and how would you wire this up to overcome any difference? You were discussing 1/2 switching. KOC shows a neat "standby"-ish switch that lifts the output cathode resistor/cap from ground. I've tried this on other amps and it works nice and quietly. Any opinions?

zevoz · Post by **zevoz** » Sat 04/22/06 7:16 pm

Boogie used a switch to disconnect 2 of the 4 cathodes as the 100/60 watt switch on Mark 2's and 3's for years. They never worried about the voltage to the other 2 tubes or the impedance mismatch. They claimed the OT was overdesigned so you could use any impedance with no problems. As far as the voltage increase you could try the zener diode on the center tap trick, possibly a DPDT switch to do both as once.

Roudyrockr · Post by **Roudyrockr** » Sat 04/22/06 7:41 pm

Just build an 18 and a 36 and be done with it. Tweak them both for great tone instead of the compromised tone of the 1/2 power switch 36. Take the 18 to the small gigs and be thankful for the relief on your back, take the 36 to big gigs and be thankful for the exposure to larger crowds. We are all collectors aren't we?

blaren · Post by **blaren** » Tue 04/25/06 2:29 am

Well it's pretty hard to argue with buying more amps. Some of us are just tightwads though and figure that if you can get 2 for the price of one (36 with half power) then woohoo...AND get an 18 watter. (Oh and chub...quit callin me a chicken...I aint yella).
I been using mine in 36W and just like Rich said, the master works REALLY well. I used it this weekend in like the smallest bar we play. No problem (by the second set). It woulda been nice though for the first set to have her screaming in 18W mode. Cluck Cluck

Is that what we were cooking at Richie? over 350? That's getting up there with the grids where they are eh? Those other numbers that Bob gets...340 with over 11V on the resistors...if I had that I'd be happy as a clam. Nothing wrong with that is there? That's with the resistor and not the choke correct Bob? Is everything else (except for the goofy cathode resistor setup you got going on lol) the same as Nik's latest layout?
I think I'm gonna put the resistor back and see what it looks like. Without the choke of course.

Post by **Plexi** » Tue 04/25/06 7:37 am

As Mark was also saying.. some 18s do run over 340 into 350 range..but the screens are usually lower.. With the 36 1/2 power switch they were very close.. to run at the higher plate voltages, on the EL84s..the screens have to be kept lower. My brother just finished a Liverpool amp,copied form Ingrid i believe it was. Ken Fischer used a 1/2 power switch on this amp,and only one he used it on,since it was the first liverpool he built.
The half power switch did work,but the samp runs a 260-0-260 PT,much different. And you don't see any half power switches on any of the other Liverpool amps. Like i said it did work,all was fine.. but i think the reason Ken didn't use it on any other amps was because it just didn't make that big of a change. I guess i'm just not a half power guy. You can get the same level with just using the MV,as you can with the 1/2 power switch.
I'm sure with some fiddling with values, you could get it to work.

Richie

blaren · Post by **blaren** » Wed 04/26/06 3:01 am

I read somewhere...wasnt it here?...about some of them peaveys I think...and others, that were making like 50W hitting EL84s' plates at around 500V but like Richie says, they had LOW screen grid voltages. I hate to say it but I think it was even 50 (claimed) watts from a PAIR not a kwawd

????
Rich..you ever gonna send me a pic of that "bass eating" filter setup (those ARE filters we're making by putting a cap in series(?) with the resistor there?) you keep braggin about?

What kind of 84s are you fellers running for your everyday use and ABUSE? I dont have a problem with the JJs at all but, those stupid "teslas?" I had in there sure gave that thing a nice musical nasalish voice. They have "round" plates and no labels. NOS. Sound great but they light up brighter than an Amsterdam ho's front window on $2 Tuesday!!! They do that in my old harmony amp too now (dont know if the POSs did before) so I put them away before I hurt myself or my amp.
So yeah. It would be nice to find something as robust as the JJs but that dont sound so trailertrash. Oh...but at trailerpark prices.

Donations

Links

Welcome

1/2 power switch data

1/2 power switch data

output impedance

I agree

One of each