EL84/6V6GT A BIT TOOOOO HOT (PLEX GAIN 20)

[TIMBO]

OK, Got some of the bugs sorted and have power beyond the standby sw. (sort of) Voltages seem fine.

NODE A)386
        B)374
        C)345
        D)306
        E)251
        F)238
EL84 pin 7 - 364 } both tubes
       pin 9 - 370 } both tubes
Across 1ohm r 54.4mA }both tubes

6V6GT pin 3 - 364 }both tubes
         pin 4 - 368 }both tubes
Across 1ohm r 49.7mA }both tubes
I'm thinking these voltages for the power tubes are a bit high,the values of the resistors on the schem are values that seem to be fairly typical for these tubes, although if the vvr was working they might be a bit lower even when at max.

What would i need to do to bring these into a safer voltage range?
The EL84 i would like to push a bit.
The 6V6 more into a bluesy feel.
Thanks

[tubeswell]

364 is nothing for a 6V6. (BFDR run them at 430). With 25mA per tube, that is about 9.1W at idle or 75%. I bet they sound really good. You could even increase the plate voltage a bit or warm then up a tad more (esp if they are JJ6V6S)

[TIMBO]

The voltage reading for the cathodes is taken across the 1ohm resistor and each tube has its own resistor and each resistor is reading (el84s) 54.2mA so do i half that to get what each tube is bias at? Thanks

[jeff]

According to the schematic the reading you're getting across each 1 ohm resistor is for each tube.

 You may want to increase the shared cathode resistor(120 ohm). Usually on schematics for a pair of EL84 you'll see a 135 ohm resistor ...But... usually the voltage on the tubes in those amps is lower. Since your B+ is higher you'll need a bigger cathode resistor to adjust the bias. The 120 ohm is not a one size fits all value. Different voltages on the tube requires a different cathode resistor.(Just to give you an example I have two SE 6V6 amps, One has 340V and a 470 ohm cathode resistor and the other has 250V and a 250 ohm resistor.)

What you want to do is measure the voltage at the plate. Then measure the voltage at the cathode. Subtract the cathode voltage from the plate voltage. This gives you the voltage the tube "sees". Multiply that by the voltage you read across the 1 ohm resistor. Your answer should be 12.

Try different resistors until you get 12. Remeasure all voltages after each resistor change and recalculate.

(plate voltage - cathode voltage) X (voltage across 1 ohm resistor)=12

Check the 6V6s too. Some people say you can go with 14 for 6V6s, but I usually go with 12.

What is the switch you're using for the power tube selector rated for? I may be wrong, and hopefully someone will speak up here if I am, but I think you can leave the PI connected to both sets of power tubes and just use a SPDT switch to lift either set of power tubes cathode resistor.

[jjasilli]

364 is nothing for a 6V6. (BFDR run them at 430).   EL-84's are often run way over-voltage.  The tonal result is a mater of some debate.  Some people feel that running them over 300V makes them sound harsh, brittle, and glassy in an unpleasant way.  If you have a variac, you can experiment with lower supply voltages.  Just keep your 6V heaters at 5V or more; that's your lower limit of "wall supply" voltage, probably about 97VAC.  You may want to lower your downstream B+ dropping resistors to maintain preamp voltage.  You could do much the same with  $9 bucking transformer.  These are handy, non-invasive ways to experiment with lower voltages.  If you like the lower voltage, there are several permanent solutions.

Trying to re-set cathode bias can be frustrating.  A larger cathode resistor does limit current, but it also pushes up plate voltage.  The higher voltage X the lower current tends to hover around the same wattage of plate diss.

[jojokeo]

Timbo,

Here's an idea I'd try to do if I was you - using a DPDT or TPDT switch if possible for doing double duty.
Setting 1 - can lift el84 cathodes AND allow full B+ to to OT and screens for 6v6s
Setting 2 - can lift 6v6 cathodes AND reduce B+ to OT and screens for el84s

Use a votage divider or large power resistor (or two) off of HT or maybe a switched 50v zener off of PT CT to lower HT for el84s. I haven't drawn it out but I think you could pull this off one way or another so a means of lowering & raising the HT can be controlled w/ a single switch AND lift one set of power tubes' cathodes. You may have to be creative so as to keep your preamp voltages relatively close in how you go about it.

[jerrydyer]

I want to buil a 6v6 plexi.. can I see your schem?
thank you

[sluckey]

http://mhuss.com/Plexi6V6/index.html

[jeff]

"Trying to re-set cathode bias can be frustrating.  A larger cathode resistor does limit current, but it also pushes up plate voltage.  The higher voltage X the lower current tends to hover around the same wattage of plate diss."

 I wasn't aware that this was a problem. So using a 120 ohm or a 470 ohm would give you the same result as far as plate dissipation?

 I wonder if this will be as much of a problem with the SS rect? I imagine with a tube rect as you draw less current the voltage drop of the tube will be less raising the B+ but do you think it may be less of a problem with SS rect.? I'm not sure, just an idea.

[jerrydyer]

oh yeah I remember that schem.  whats the scoop on just putting 6v6's into a 2204 circuit? what do I need to change. I know they are pretty rugged little bastards.

[sluckey]

Lower the B+ to the 6V6s, change the bias, and maybe diddle the PI. Probably easier to just copy Marks PI, PA, and PS but replace the Plexi preamp with a 2204 preamp.

[jojokeo]

"Trying to re-set cathode bias can be frustrating.  A larger cathode resistor does limit current, but it also pushes up plate voltage.  The higher voltage X the lower current tends to hover around the same wattage of plate diss."

 I wasn't aware that this was a problem. So using a 120 ohm or a 470 ohm would give you the same result as far as plate dissipation?

Negative! That's like saying, "just throw any resistor in there, it's all the same result."

It absolutley makes a difference and is not hard or frustrating at all to bias a self-bias/cathode bias amp. All it takes is going through the exercise to install, measure, and repeat if necessary a couple times. No biggie really.

The earlier statement is basically true due to the two factors being indirectly proportional is all. The lower the plate voltage, you will need a lower value resistor to allow or set more current to achieve the proper Pdiss. The higher the voltage, then the higher value resistor that's needed to limit more current & prevent over-shooting Pdiss and red-plating. But, this isn't "hard" to do or find the correct value for your situation. It's very easily done.

[PRR]

> using a 120 ohm or a 470 ohm would give you the same result as far as plate dissipation?

No. Small resistor changes give very-small results. Your example 4:1 resistor change tends to cut current in half. OTOH with a saggy supply, half current may cause a significant rise in plate voltage, so 1:4 resistor is still more than half dissipation.

[jeff]

That's like saying, "just throw any resistor in there, it's all the same result."
OK that was confusing. The way it was worded I thought that's what he was trying to say.
 
 So Is this what TIMBO needs to do to solve his problem?

 Measure the voltage at the plate. Then measure the voltage at the cathode. Subtract the cathode voltage from the plate voltage. This gives you the voltage the tube "sees". Multiply that by the voltage you read across the 1 ohm resistor. Your answer should be 12.

Try different resistors until you get 12. Remeasure all voltages after each resistor change and recalculate.

(plate voltage - cathode voltage) X (voltage across 1 ohm resistor)=12 

or else lower the voltage(maybe use a 5Y3???? check the current draw)

[TIMBO]

Hi guys, I tried to post a reply this morning but the computer decided it did not want to work   and you guys have been posting galore,so what i was going post doesn't matter.I need to get the VVR up and running to see if this might reduce the HT voltage even if is at max.I took readings of the PT when loaded a i was getting 325v on both HT leads, which gave me about 405v after rect, i was surprised at this as it is a 290-0-290 PT   So is the possibility of using a zenner on the PT CT a way of fixing this,the preamp voltages at the moment are good but i am able to adjust the dropping resistor to compensate this. WOW information overload  

Tubes:-
JJEL84 / Tung-sol 6V6gt

1)Will these be happy with a plate voltage of Approx. 364v ?
2) i'll will use the calculation that jeff has supplied to see what i get with the plate voltage as is.

Thanks
Just did a check on the PT i'm getting 315v at the PT leads and 420 after rect

[sluckey]

I took readings of the PT when loaded a i was getting 325v on both HT leads, which gave me about 405v after rect, i was surprised at this as it is a 290-0-290 PT

Check your line voltage. Is it high too?

So is the possibility of using a zenner on the PT CT a way of fixing this

Why bother with a zener? You got VVR which does the same as a zener. VVR is even better because you can adjust how much voltage is wasted across the MOSFET rather than rely on wasting a fixed amount of voltage on a zener. Just think of the zener as a VVR without the pot.

[jojokeo]

Why bother with a zener? You got VVR which does the same as a zener. VVR is even better because you can adjust how much voltage is wasted across the MOSFET rather than rely on wasting a fixed amount of voltage on a zener. Just think of the zener as a VVR without the pot.

Yeah that's correct and I was just throwing something as a possibility out there, but I was thinking of the amp having a switch that will automatically adjust for either set of tubes which would correct everything at full power. Then the VVR simply adjusts everything down from there.

The other way you'd have to remember to turn the VVR down by a certain amount before you flipped the switch to run the el84's. You may remember the first month or two but down the road this could easily be forgotten.

[tubeswell]

The voltage reading for the cathodes is taken across the 1ohm resistor and each tube has its own resistor and each resistor is reading (el84s) 54.2mA so do i half that to get what each tube is bias at? Thanks

Sorry I'll read that again.

I assumed, because of the amount of current that you were measuring, the you were the measuring the tube current in both tubes combined through a single shared biasing resistor. But I see you meant you had ~50mA in each tube 50mA is way too high for a 6V6 running at ~360V. Its about 2 x as high as it should be (which is why I made the wrong assumption sorry)

[plexi50]

This has helped me a lot latley with SE amps in determining the cathode bias resistor needed

http://www.webervst.com/tubes1/calcbias.htm

[TIMBO]

Thanks plexi got that on my desktop but its a little technical for me at the moment but i do understand the tables at the bottom is a good guide ( the values shown are taken across the 1ohm resistor or at the socket pin)

I plan to bypass the VVR cause the mosfet is a dead duck.To get another is $25 and a week wait and the possibility of using PPIMV could a goer.

I do know that the local power supply can go up and down at certain times of the day, so i'll take some readings later.
I do have room on the circuit board to be able to have separate cap/resistors for each power tube, could this help?

Thanks

[kagliostro]

I think you can save some money

http://www.el34world.com/Forum/index.php?topic=9623.msg87366#msg87366

http://it.rs-online.com/web/p/products/5411089/?searchTerm=IRFPE50&relevancy-data=636F3D322664663D5926696E3D4931384E4B6E6F776E41734D504E266C753D6974266D6D3D6D61746368616C6C7061727469616C26706D3D5E5C772B2426706F3D313326736E3D592673743D4D414E5F504152545F4E554D424552267573743D495246504535302677633D424F544826

about the automatic change of VVR regulation I think you can play with a resistor in series to the pot circuit and a switch in parallel to short the resistor, so when you act the change of tube at the same time you change voltage to the tube (use a DPDT Switch for that, so one side acts the cathode swap, the other acts the resistance change)

Kagliostro

[TIMBO]

OUTSTANDING Kagliostro , I have a local RS store here a Brisbane, Australia. Thanks

[kagliostro]

Very happy to know you have RS store in Australia

however my intention was only to show that there are other mosfet that are cheaper than the NTE2973 and can be used safety

Kagliostro

EDIT: if you don't know, RS has also this: RS Stock No. 433-703

[TIMBO]

OK, This is where i am at:-

         res          across cath res    across 1ohm res   plate - cath     watts/tube

EL84  120r               13.2v                  56.2mA              369v           19w
        250r               16v                                             373v           11w
        470r               18.4v                  19.7mA              387v            7w

6V6   250r               25.3v                  51.4mA               369v          22w
        407r               30.8v                                            357v          11w
        560r               32.8v                  28.9mA                361v          10w


If i have used the weber bias cal properly by increasing the r the wattage goes down but the plate voltage goes up.This exceeds the max for the tube plate voltage. How does these values sit with you guys. Thanks

[sluckey]

PA_max for EL84 is 12 watts. I would not exceed 90% (10.8 watts). I'd like to see the numbers for a 330Ω cathode resistor.

PA_max for 6V6 is 14 watts. I would not exceed 90% (12.6 watts). I'd like to see the numbers for a 330Ω cathode resistor for this one too.

[jeff]

But the current you're reading is for the plate and the screen. If the screen current is about 10% of the plate current then

[sluckey]

But the current you're reading is for the plate and the screen. If the screen current is about 10% of the plate current then

That's right. I don't know that the screen is 10% of the plate, do you? If you must be exact, then you must measure screen current or calculate it from the voltage drop across the screen resistor. By ignoring the screen current, you err on the side of caution, ie, the calculated PA_dis will be slightly higher than the actual PA_dis.

[TIMBO]

New readings :-

Wall outlet - 250v
PT HT - 318v each leg
Rectifier - 395v

             r      across cath    plate - cath        across 1ohm r     watts
6V6     330r       27.8v            354v                42.5/41.9mA      14w
          390r       29.4v            359v                38.1/37.4mA      13w

EL84   330r       17.2v             378v                26.5/26.2mA       9w
         390r       17.8v             383v                23.1/22.7mA       8w

These look like they are in the ballpark 

[jeff]

"That's right. I don't know that the screen is 10% of the plate, do you? If you must be exact, then you must measure screen current or calculate it from the voltage drop across the screen resistor. By ignoring the screen current, you err on the side of caution, ie, the calculated PA_dis will be slightly higher than the actual PA_dis."

Thanks Sluckey, You've made me question my thinking. I may have gotten some bad info. I was told how to use the tube charts to figure the screen current vs. plate current ratio of a tube but that may have been a myth/rule of thumb. I'll investigate that further, but thanks for bringing that up.

Is this what you'd use if you wanted to be presice?:

[(plate voltage-cathode voltage)/(voltage across 1 ohm resistor)]-(voltage drop across screen resistor/value of screen resistor) X 90%= 12 ooops

[jojokeo]

I forget where I got this info (maybe PRR or Merlin?) but to get a good idea of screen current divide plate current by 6.5 to get you close.

[sluckey]

Is this what you'd use if you wanted to be presice?

No.

1.   I'd calculate the real plate voltage by subtracting the measured cathode voltage from the measured plate voltage.

2.   Then calculate or measure cathode current with 1Ω resistor or bias probe.

3.   Then calculate screen current by measuring the voltage drop across the screen resistor and dividing by the value of that resistor.

4.   Now calculate plate current by subtracting screen current from cathode current.

5.   Finally, multiply plate voltage times plate current.



Side note... We've all seen the 1Ω trick used to measure cathode current. But did you know that you can use the 1Ω trick to measure current in other places?
For example...

1.   Put a 1Ω resistor in the plate circuit to directly measure plate current without having to deal with screen numbers.
2.   Put a 1Ω resistor in the B+ line to directly measure B+ load current.
3.   Put a 1Ω resistor in series with the AC line fuse to measure the total AC line current. (Need a good true RMS meter).

The 1Ω resistor is not a one trick pony. But using it in the examples above require 2 handed probing or clip leads. For those that subscribe to the "one hand behind your back while probing" theory (I don't, I use the know what you're doing and be careful theory) or you don't have steady hands, this may not be a good idea.

[jojokeo]

The 1Ω resistor is not a one trick pony. But using it in the examples above require 2 handed probing or clip leads. For those that subscribe to the "one hand behind your back while probing" theory (I don't, I use the know what you're doing and be careful theory) or you don't have steady hands, this may not be a good idea.

+1

I will suport my hand against the chassis on occasion so that the probe doesn't slip off where I've got it and I've been guilty of touching things like a standby or power switch from time to time 

[jjasilli]

The 1Ω resistor is not a one trick pony. But using it in the examples above require 2 handed probing or clip leads. For those that subscribe to the "one hand behind your back while probing" theory (I don't, I use the know what you're doing and be careful theory) or you don't have steady hands, this may not be a good idea.

+1

I will suport my hand against the chassis on occasion so that the probe doesn't slip off where I've got it and I've been guilty of touching things like a standby or power switch from time to time 

I'm too clumsy!  I use clips on my probes.  Otherwise I always short-out something with the probe tip.  With probe clips I can attach one at a time.  Ground side always goes first, except for floating measurements which are always live on both sides.  Also I like to use more than one meter to keep an eye on, say, plate voltage while measuring something else like current across a resistor.  I like to use VTVM's to monitor high DC voltage, and digital bench meters for floating measurements.  Hand held meters are fine if you can find it (the meter, not your hand!)