Bias Adjustment for Crate Vintage Club 50

[punkykatt]

Hey Guys, i have a mid 90`s Crate Vintage Club 50 on the bench in for cleaning a service. Everything is working fine except the bias seems to be  hot (but not red plating) for a fix bias non adjustable (4) EL84`s  at  85 to 100% mpd.  Schematic calls for plates to be 391vdc, I get 366 to 371vdc depending on line voltage. At the junction of the 220k`s schematic  calls for -15.6vdc,  I get -14.6vdc.  I tried three sets of quads, they all read in the same ballpark.  In the bias circuit the range resistor R77 is only 100 ohm.  How can i cool the bias down? Or was this amp designed to run hot?  Thanks for any input.  Punky

[sluckey]

In the bias circuit the range resistor R77 is only 100 ohm.

That doesn't give much room to run cooler.  What is the measured -V at TP27? Your bias voltage can never be more than that.

You can replace jumper J03 with a 100Ω resistor to run cooler. My Magnatone M10A has a non adjustable -16V for the grids and a 100Ω cathode resistor.

[Willabe]

What is the measured -V at TP27? Your bias voltage can never be more than that.

Depending on if that -dcv is large enough can't he increase R78?

(I take it he may only need a few more -dcv?)


                Brad   

[sluckey]

Depending on if that -dcv is large enough can't he increase R78?

Sure. You might get a few more millivolts. Look at the whole voltage divider... 100Ω and 15,000Ω. R78 already has 99.3% of the voltage available at TP27. 

[stratele52]

A bigger value for  R 78 will do the job for a colder bias and if needed smaller or no R77

[HotBluePlates]

... Schematic calls for plates to be 391vdc, I get 366 to 371vdc depending on line voltage. At the junction of the 220k`s schematic  calls for -14.6vdc,  I get -15.6vdc. ...

Focus on screen voltage and bias voltage, as those change the plate current. Are the screens also lower voltage than the schematic value?

You quoted lower-than-schematic plate voltage, and higher-than-schematic bias voltage... that would cool the tubes off, right? So I'm guessing the amp was designed to run this hot (schematic bias is actually -15.6v like you got, so no reason to tinker?).

[Willabe]

R78 already has 99.3% of the voltage available at TP27.

That's already about it then.


               Brad         

[punkykatt]

My quote was wrong on the TP 17 junction of the 220k resistors. My reading is -14.6vdc schematic calls for -15.6vdc , my bad.  Also the screens are running low as well, my readings are  356vdc, schematic calls for 381vdc.   The highest negative voltage reading upstream of R77 is -15.16vdc.  Im thinking this amp was made to run hot?

[HotBluePlates]

Sounds right.

Bias voltage has dropped (higher plate current), but so has the screen voltage (lower plate current).

The EL84 has a triode Mu of 19. So the control grid is 19 times more effective at controlling plate current than the screen (plate voltage is near-meaningless until it is below ~50v). Screen voltage change is 381v-356v = 25v. Grid bias change is -15.6 - -14.6v = 1v.

G2 change  -->25v/19 = ~1.32
G1 change  --> 1v

The tubes should actually be running cooler than stock, because the screen volts change cooled off plate current more than the grid bias change heated up plate current.

[stratele52]

... Schematic calls for plates to be 391vdc, I get 366 to 371vdc depending on line voltage. At the junction of the 220k`s schematic  calls for -14.6vdc,  I get -15.6vdc. ...

Focus on screen voltage and bias voltage, as those change the plate current. Are the screens also lower voltage than the schematic value?

You quoted lower-than-schematic plate voltage, and higher-than-schematic bias voltage... that would cool the tubes off, right? So I'm guessing the amp was designed to run this hot (schematic bias is actually -15.6v like you got, so no reason to tinker?).

Voltage on schematic are good IF you use same wall outlet voltage than desing need AND same output tubes as original from the builder . Anyway  I see sometimes amps bias too hot direct from the store .


Schematic help but did not say all , you need some knowledge too.

[PRR]

> Or was this amp designed to run hot?

Yes.

50 Watts out of quad-EL84 is *very* bold engineering.

If an airplane, or a bridge, were designed that bravely, you'd never want to get on it.

EL84 are cheap. Idle them at 90%, 99%, 115%, what-ever, and replace them when they go bad.

[smackoj]

I did not look at the schemo so maybe this is a 'dumb' question, but why not convert to adjustable fixed bias and dial the pup where the player likes the sound best? I understand this would be an additional cost to the owner but seems worthwhile to this old fogey?

 

[sluckey]

I did not look at the schemo so maybe this is a 'dumb' question, but why not convert to adjustable fixed bias and dial the pup where the player likes the sound best?

The challenge is that he would like to run cooler. To do that he would need to increase the bias voltage. The problem is that he is already at maximum bias voltage and there is no more available. So, adding a pot will only allow him to adjust for even hotter operation which ain't what he wants to do.

Most all EL84 guitar amps I'm familiar with seem to run the bias very hot. Heck, the AC-15 I just built idles at 14.75W per tube!

[stratele52]

I did not look at the schemo so maybe this is a 'dumb' question, but why not convert to adjustable fixed bias and dial the pup where the player likes the sound best? I understand this would be an additional cost to the owner but seems worthwhile to this old fogey?

 

+1

[stratele52]

  . The problem is that he is already at maximum bias voltage and there is no more available. 

 

Sure more voltage is avalaible . To bad people do not read ( or understand  ? ) my post April 29 , 4:42 pm


"   A bigger value for  R 78 will do the job for a colder bias and if needed smaller or no R77  "

[sluckey]

  . The problem is that he is already at maximum bias voltage and there is no more available. 

 

Sure more voltage is avalaible . To bad people do not read ( or understand  ? ) my post April 29 , 4:42 pm


"   A bigger value for  R 78 will do the job for a colder bias and if needed smaller or no R77  "

You don't know what you're talking about! If you won't believe me maybe you can look at the chart of test points on the schematic. Maybe you'll believe that!

[stratele52]

  . The problem is that he is already at maximum bias voltage and there is no more available. 

 

v
Sure more voltage is avalaible . To bad people do not read ( or understand  ? ) my post April 29 , 4:42 pm


"   A bigger value for  R 78 will do the job for a colder bias and if needed smaller or no R77  "

You don't know what you're talking about! If you won't believe me maybe you can look at the chart of test points on the schematic. Maybe you'll believe that!

Just I'm saying is I read post ( and try to understand I'm french Canadian ) I give the way to have more negative voltage , bigger value for R78. 15 k drop to ground more voltage than let say a 30 K . Sure if you put 30 K you'll have colder bias .


Some say to replace R77 the 100 ohms , I said maybe but work on R78 first .


I do not read Punkykatt try a new vallue for R78 and said it don't work . I'm wrong ?

[sluckey]

  . The problem is that he is already at maximum bias voltage and there is no more available. 

 

v
Sure more voltage is avalaible . To bad people do not read ( or understand  ? ) my post April 29 , 4:42 pm


"   A bigger value for  R 78 will do the job for a colder bias and if needed smaller or no R77  "

You don't know what you're talking about! If you won't believe me maybe you can look at the chart of test points on the schematic. Maybe you'll believe that!

Just I'm saying is I read post ( and try to understand I'm french Canadian ) I give the way to have more negative voltage , bigger value for R78. 15 k drop to ground more voltage than let say a 30 K . Sure if you put 30 K you'll have colder bias .


Some say to replace R77 the 100 ohms , I said maybe but work on R78 first .


I do not read Punkykatt try a new vallue for R78 and said it don't work . I'm wrong ?

DID YOU EVEN LOOK AT THE CHART???

[stratele52]

Witch chart  ?


I look the schematic

[sluckey]

chart on schematic

[stratele52]

chart on schematic

Oh that !..nothing usefull for colder Bias.






Replace R78


 

[Willabe]

The chart on page 2, top right corner lists TP27 as -15.7dcv. That's all you can get.

Punkykatt is getting a reading of -14.6dcv.

The PT -bias supply can only give another -1.1dcv at the most, that's not enough.


            Brad     

[punkykatt]

Thanks for all the replies everyone.  Replacing R78  with what value? will give me that one -volt increase from the PS and how much will it change (cool) the bias? Its a PCB amp and not easy to swap R78.

[sluckey]

Thanks for all the replies everyone.  Replacing R78  with what value? will give me that one -volt increase from the PS and how much will it change (cool) the bias? Its a PCB amp and not easy to swap R78.

Since it's not easy to swap R78, just put a jumper (gator clip test lead) across R77. That will give you the absolute most negative voltage you can get from that bias supply. You will not damage anything doing this. Please measure the bias voltage with and without the jumper so we all may understand.

[sluckey]

chart on schematic

Oh that !..nothing usefull for colder Bias.

Replace R78

Let me try to explain it again. From the chart, TP27 is -15.7VDC. That's straight out of the rectifier. That's the most the bias voltage can ever be. That voltage is applied to a voltage divider consisting of R77 and R78. The voltage across R78 is the negative voltage that will be sent to the grids of the output tubes.

R77 is 100Ω and R78 is 15,000Ω. It should be obvious that most of that -15.7VDC will be present across R78 and that very little will be dropped by R77. But let's be sure. Doing the math shows that -15.6V is across R78 and only -0.1V is across R77. So, 99.4% of the total available -15.7VDC is already being sent to the output tube grids.

Now let's say that R77 is reduced to zero ohms. R77 now drops zero volts and the entire -15.7VDC appears across R78. This is only an increase of 0.1V and that small increase is likely to have no effect on biasing the output tubes cooler.

Now let's go the other way, your preferred method. Leave R77 at 100Ω and let's increase R78 to 100K. We still have -15.7VDC applied to this voltage divider. Doing the math again shows that now the voltage that appears across R78 has increased to -15.68. Wow! The bias voltage increased from -15.6 to -15.68. Not likely to make the output tubes run any cooler. Even as you increase R78 to infinity, the bias voltage can never be more than the voltage at TP27.

Your idea of increasing the value of R78 to increase the bias voltage to run the tubes cooler would be valid ***IF*** R78 was not so much bigger than R77. For example, if both resistors were equal value then half the -15.7VDC would appear across each resistor. In that case, you could increase R78 (or decrease R77) and get a big enough voltage change to actually do something.

[punkykatt]

OK, today the bias voltage(my line voltage varies) at TP30(junction of R77, R78) is: without jumper across R77 -14.90vdc,   with jumper across R77 -14.99vdc =  .09vdc difference.   Crate made this a hot running amp.  Thank you everyone for all the replies.  Punky

[PRR]

Coming late to the discussion....

> Sure more voltage is avalaible

Trusting the schematic....

The PT winding, diodes, and C35 are approximately a 30 Ohm power source.

R77 adds 100 Ohms.

Into 15K load, the total 130 Ohms series resistance leaves 0.991 of the unloaded voltage.

IOW, the most bump possible is 1%.

If tubes are hot, 1% bias change won't cold them down enuff to notice.

[PRR]

Sure more voltage is available. We can build a voltage-doubler off the existing winding. Or cap-couple off the 240VAC winding to jack-down a negative DC voltage. Either can easily give ~~twice the voltage we need, which is easily trimmed-down to any reasonable bias.

However: "on the bench in for cleaning a service."

The customer is not(?) complaining about hot tubes.

If the bias were obviously "wrong", Punky would "have" to look into that because if it burns-up right after "cleaning and service" the customer would be mad and Punky would be sad.

But the numbers are essentially "AT" the service drawing values. So close that any deviation may be parts tolerance or even factory changes (the 5%-low B+ with all else within 1% suggests that Crate learned they had built a toaster and wound-down the B+).

I would not dink with the bias without talking to the customer. And I might just ask "Has it always run this hot?" If it has been that hot since the mid 1990s, 20 years, then that's the way it is, and no urgency. (Yes, if he says it used to run cooler, I'd try different tubes, then ponder a bias-boost.)

[stratele52]

Sorry 


You are all right about voltage , no more is avalaible

[smackoj]

I did not know that 'Crate built a toaster' but after trying to make a Crate TD50C sound better than a Kenmore washer and dryer combo, I would not be surprised to learn that some elec. engineer students are the bulk of Crate's R and D dept. Whiz kids with a slide rule and a family size box of 'Good and Plenty'.   

[Jack1962]

I have worked on a few of these amp , yes they are designed to run hot , the go thru those EL's pretty fast if you use the amp . You have to push those EL's hard to get 50 watts so Crate in there wisdom(roflmao) biased them right on the edge of meltdown.

[bnwitt]

What about replacing the full wave rectifier in the B- circuit with a bridge rectifier?  Or would that be worse I think?

[HotBluePlates]

What about replacing the full wave rectifier in the B- circuit with a bridge rectifier?  Or would that be worse I think?

That could work. Or a simple voltage doubler could be used. It seems Crate designed the amp to run as hot as it does, though.

[sluckey]

Pencil it in and see what happens. You're gonna pop some diodes and if they stay shorted you'll start blowing fuses or even worse.

There is already a FWB in the circuit. And the center tap is used so the bridge puts out negative and positive voltages. If you replace the two diodes in the negative half of the bridge with another FWB, you still gotta leave the center tap connected for the positive output . Leaving the CT connected will insure that the diodes in the new bridge will pop.

There are ways around this including using another transformer strictly for the bias rather than modify the negative supply that doubles as a bias supply and a supply for some solid state stuff. But you will quickly run into "why bother" territory.

I still think the easiest way to make this amp run cooler is to add a 100Ω resistor under the cathodes. Even easier to accept it as a toaster.   

[HotBluePlates]

Pencil it in and see what happens. You're gonna pop some diodes and if they stay shorted you'll start blowing fuses or even worse.

There is already a FWB in the circuit.

Duh... Thanks for pointing that out (might help if I go back and look at the schematic since it's been a while).

Yes, accept the toaster. Or follow your earlier suggestion to source raw a.c. for the bias circuit from a cap to the high voltage secondary, then knock down as appropriate.

But this amp is what it is.

[punkykatt]

Its a toaster for sure, but none of the el84`s in the  three sets of quads  that were used for testing were red plating.  The amp does have a 0.5A fuse  down stream of the standby switch in case of  plate melt down.  Both the AC line voltage  and the HT fuses are internal on the amp????? You have to pull the chassis to get to them.   NOT a good amp to gig with????? It`s a combo with 3 tens and weighs a ton.  What were they thinking????  OH Yeah, they sell you a back up amp.

[Willabe]

What were they thinking????  OH Yeah, they sell you a back up amp.

       

[1fbfanhere]

I know it's been a while since the original post, but I feel I should comment. I purchased a Crate Vintage Club 50 3X10 for $75. All of the output tubes were blown. So I took it apart to see what was going on inside. I am far from a professional. However I have a good soldering iron and I know how to use it. I am also good at finding info on the net. I feel Sluckey is absolutely correct. What he didn't know was that the JO3 is made to have a resistor and right next to it is an empty C29. I found a schematic for a Club 33 and it appears to be the same board. In the JO3 is a 10watt 80ohm resistor and the C29 has a 220uf 60 volt electrolytic capacitor. I had to replace three 47uf 450 volt C14, C15, C25 and the one 10000uf 16volt C57.  Everything else was good. I then put a 100ohm 10watt at JO3 and a 220uf 60 volt at C29. It is still runs very hot, but it works great. Sounds awesome. I have also plugged the amp into a variac running at 100 volts. There is a lot of dispute about whether you should do this or not so I am not recommending this to anyone. To me it appears they had to run it very hot to get the 50 watts out of four EL84's. As long as you keep it well ventilated it works, but still go through tubes faster than normal. However, if you have it up against the wall, in a hot corner, put something on top, etc. it easily over heats. So if you have one and haven't had any problems, I would put a fan inside. If you have had problems, you could replace the bad components and put a fan inside. I still think you will go through tubes fast that way.  I will update in a few months to see if what I did actually helps. Rock on.