Metal film (blue) vs. film (tan) resistors

[dude]

Getting ready to build another Allen Old Flame. I had bought one of David's first kits in the nineties, a great amp. I used the carbon comps that came with the kit, eyelets and board screwed to the chassis just like the Blackface's of the mid Sixties. Never had a problem with the amp, single channel, MV and reverb. The amp really came alive when I got hold of a used good matched pair RCA 6L6 Blackplates for $30. Used a Magic Parts multi-tap OT, probably 50 watts, rather than the stock David Allen OT at that time, only 4 ohm secondary.


This time I have a raised bd with eyelets and and was thinking about using film resistors but the leads are too short for my already made bd. so I was thinking about the blue metal film resistors as the leads are thicker and longer.


Would there be any difference in tone with these blue metal film resistors? I've used them for repairs and mods but never a new build. I assume Doug sell them, haven't looked yet.


Also, what benefit would using 1 watts rather then half, where their called for? I assume nothing...? 


al

[sluckey]

I've been using 'blue' metal film exclusively for a few years now (unless I just can't find the appropriate value in mf. I like the fact that they measure what they should measure. Blue is my favorite color. I think they sound kinda bluesy with a bit of lite metal thrown in the mix. I could go on and on! 

Also, what benefit would using 1 watts rather then half, where their called for? I assume nothing...? 

Good for the economy.

[bnwitt]

I just finished building 3 Hoffman AB763 Deluxe Reverb amps.  The first one was all CC resistors except the power rail resistors where I always use metal oxide.  The second amp was a mix of metal film(blue), carbon film(tan) and the metal oxide.  The third amp was MF and MO.  All three sounded fantastic but the CC amp had some background hiss at high volume settings initially which almost disappeared after several hours of playing.  The last amp had 1 watt MF resistors and 1 or 3 watt MO resistors. It is very quiet on idle.  I'll probably never use CC resistors again.

[dude]

I think they sound kinda bluesy with a bit of lite metal thrown in the mix. I could go on and on! 

Where's the lite metal ?


Well, question answered, accurate, longer, thicker leads, low noise compared to CC, half watts are fine .




al

[sluckey]

Where's the lite metal ?

It's under the lite blue paint. 

[labb]

Do you guys ever consider a resistor's voltage rating? I'm not referring to its power rating (ie. watts) but its voltage rating in volts..Most 1/2 watt resistor's data sheet shows a working voltage rating ( 350 volts) with a maximum over load voltage ( 600 volts). So if I have a B+ of say 460 volts do I need to be sure that the voltage rating is over 460 volts...Seems like a simple question but most that I have asked this of always go back to a power rating not the voltage rating. example would be a xicon  220k, 3 watt metal oxide. Data sheet shows 350 w.v. with a over voltage  of 600. Could I use this resistor with the B+ of 460 volts.

[sluckey]

The voltage rating refers to the voltage measured ***ACROSS*** the resistor, not the voltage from one end to chassis ground.  For example a screen resistor is connected to a 400V B+ node and the screen voltage at the tube socket pin is 395V. There's only 5 volts across the resistor.

[bnwitt]

I'm using 1 watt 500V working 750V overload.  And what Sluckey just said.

[PRR]

It's rare to find a plate resistor with more than 200v across it.

[labb]

You guys certainly know more about this than I do but,,,but,,but  I think that the voltage rating of resistors is practically the same as the voltage rating of wire...It has to do with the voltage applied to the resistor not the voltage across it. We get away with those that are rated at 350 volts because they are not physically in contact with the case or another low potential item. The only situation I get concerned about is the balancing resistor across caps when you put them in series to increase the voltage rating of the caps. They are generally on the cap. housing.    JMO

[bnwitt]

http://music-electronics-forum.com/t145/

[labb]

And from another site:

• Voltage rating : as with wire insulation, make sure this will handle the voltages you care about. In most resistors, you only care about the voltage drop across the resistor, but if the resistor is up against, or can possibly touch, another conductor (such as a ground lug!) or another resistor, you have to worry about the maximum potential between the resistor and the other conductor or resistor.

[bnwitt]

So Leo Fender, Jim Marshall and a virtual plethora of other amp builder designers died before they learned this? And you're worried about a 350 volt resistor touching a capacitar can while both of them are exposed to 2.0 volts above ground?

[labb]

Mostly I don't worry about it at all. I just use the 1 watt or larger 350 volt rated ones and go ahead. Why do you use the ones rated for 500 volt and 750 volt overload.

[bnwitt]

I use them because they're 1 watt rated and that typically reduces noise in a 1/2 watt application.  The down side to it is it's going to take a higher current to burn them out in the event of a tube short and that may cause more collateral damage.  350 working voltage and 500V peak has been used in tube amps for years and many of those resistors installed back in the 50s are still working.


Here is a definition for resistor working voltage off of an EE site:


 It is the maximum voltage that the insulation strength or the internal construction of a resistor can stand. This rating comes into consideration for resistors that have high ohmic values so their wattage or dissipation ratings are not exceeded before insulation breakdown.An example is a 10 Mohm 1 W resistor.
The wattage rating is not exceeded with an applied voltage less than about 3100 V, but usually normal resistors are rated at 500 V. This means that the dissipation rating cannot be exceeded before the insulation breaks or a flash-over occurs.

So 3100V divided by 10,000,000 equals 0.00031 amps.  Then, 3100V times 0.00031A equals 0.961 watt or within the 1 watt rating.

[labb]

My feeble mind seems to remember that there is an equation that calculates the maximum voltage allowed. It is something like the square root of power x resistance. I am too far removed from the theory to remember for sure and I am too lazy to try to look it up..Anyway thanks for the discussion. Lets get back to building and repairing amps. I've got two amps I have got to look at when I find the time (flipping a house for wife right now). One is a Crate Blue Voodoo the other is a Marshall TL60. The Marshall will be fun. It was sent to a Marshall shop and they could not get it to fail. I have watched it and it will lose filament voltage for a short time then it comes back. Sounds like a bad joint but I haven't found it and the shop didn't. I will be begging for help on these two. The Blue Voodoo was a wreck when I got it. The bias circuit was completely gone. Components and solder joints. That thing is auto self biasing. got it up and running. Now power fades in and out.

[HotBluePlates]

My feeble mind seems to remember that there is an equation that calculates the maximum voltage allowed. It is something like the square root of power x resistance. ...

That's the formula to find the voltage across a resistor (or load) when resistance and power are known.
Power = Volts²/Resistance, so
Volts = √(Power * Resistance)

Do you guys ever consider a resistor's voltage rating? ...Most 1/2 watt resistor's data sheet shows a working voltage rating ( 350 volts) with a maximum over load voltage ( 600 volts). ...

I only ever thought about it once I read for the first time that resistors have a voltage rating.  Then I realized what PRR and Sluckey said makes sense.

I haven't had any amps burst into flames yet...

[Willabe]

I haven't had any amps burst into flames yet...

Uh-oh, now you left the door open for Jim. 

[bnwitt]

You might have a cracked solder joint on the DC heater circuit rectifier BR1 pins:

http://music-electronics-forum.com/t26081/

On every JCM 2000 that has come though my shop I have removed the PCB from the amp and taken it outside in the bright sun while wearing my head magnifier to look for cracked ROHS lead free solder joints.  I have usually found numerous ones and almost always on the BR1 rectifier.  That rectifier is mounted right down on the board and the heat cracks the joints.  Installing a new one with longer leads mounted above the board a little so the heat can dissipate helps a lot.  I'd start there for sure but it would be good measure just to run a soldering iron over any suspicious joint while you have the board out.  I believe the TSL60 is similar to the TSL boards shown below
Barry

[labb]

Thanks for the info. Would a problem there also take out the power tubes filaments. Looking back at my notes one thing I noted was that the filament a.c. voltage was 5.6 volts..That is pretty low

[bnwitt]

Well I would disconnect the PT heater wires and check the voltage there to make sure it's not an issue with the board itself.  Marshall sells the complete main circuit board for these amps for a reason and that is because the ROHS wave soldering process sucks.  I've seen joints barely coated with solder, cracked solder joints etc etc.  I'm not a big fan of tube sockets soldered to a PCB either.  This amp is just bad construction all the way around.  I usually suck out the lead free and replace it with Tester lead solder any where there is heat build up like the rectifier, sockets etc.

You know it's pretty darned sad an authorized Marshall service center doesn't know all of this!  It's common knowledge in the repair industry.  I've tried to become an authorized Marshall service center but they want you to have a store front.  Have they never heard of Amazon for cryin' out loud.

[bnwitt]

There are quite a few issues with the JCM 2000 series of amps.

[PRR]

There is good reason to use 1+W resistors in plate feed of guitar amps.

Normally we find 100V-140V across 100K so 0.1W to 0.2W.

You hit a chuck-hole with the Econoline, the tube guts collapse to a short. Plug-in and now you have zero across the tube and 300+V across the 100K. Dissipation is now 0.9 Watts. At 400V supply, 1.6 Watts. A half-Watt part will die while you stand there wondering why it won't work. A 2W part will take the over-stress for months. You still have to replace the tube, but you don't have to replace the resistor. With luck/foresight, you can replace the tube before the drummer sets-up and nobody knows you had a problem; a smoked resistor normally kills the gig.

[2deaf]

You hit a chuck-hole with the Econoline, . . .

You hit a chuck-hole with the Econoline (spell check don't like it) and you will be lucky to get to the gig at all.  They had a really trick steering system that would lurch the van into oncoming traffic with just a small dip.  And how about those perpetually loose king pins.  My '56 Jeep breaking into a shimmy was annoying, but at least it went in a straight line. 

[HotBluePlates]

And from another site:

• Voltage rating : as with wire insulation, make sure this will handle the voltages you care about. In most resistors, you only care about the voltage drop across the resistor, but if the resistor is up against, or can possibly touch, another conductor (such as a ground lug!) or another resistor, you have to worry about the maximum potential between the resistor and the other conductor or resistor.

A film resistor is made by depositing a metal-resistive or hydrocarbon mist onto a (usually ceramic) rod.  If endocarps & leads are fitting to the resulting device, the resistance will be quite low.  To increase the value of the resistance, the path-length for current between ends is increased by etching away some of the surface material to form a long spiral, either by grinding or with a laser.

The resistor's working voltage is a voltage above which there could be arcing from one part of the spiral to an adjacent part, reducing the effective resistance & increasing current (possibly damaging the resistor).

The outer coating applied over the resistive spiral is a completely separate dielectric to withstand voltage, and is quite a bit thicker than the space between adjacent parts of the resistive spiral.  The dielectric strength of that outer coating is what you're thinking of when you worry about the voltage from the resistor to a nearby grounded object, and intuitively you can see it must be higher than the working voltage of the resistor because the thickness/distance is greater than the distance between adjacent parts of the resistive spiral.

Similarly, the working voltage of a capacitor is the dielectric strength of the insulator between plates of the cap; the insulation enclosing the capacitor is almost always thicker.

My feeble mind seems to remember that there is an equation that calculates the maximum voltage allowed. It is something like the square root of power x resistance. ...

That's the formula to find the voltage across a resistor (or load) when resistance and power are known.
Power = Volts²/Resistance, so
Volts = √(Power * Resistance)

Remember this?  √(0.5w * 100kΩ) = 223v, and there's always that old rule of thumb about using a resistor rated at least double the normal dissipation.  So as PRR pointed out earlier, this implies we usually only have ~100v across a typical 100kΩ plate resistor.

Whether any of us pay attention or not, √(Power * Resistance), which is associated with the resistor's wattage rating, usually takes care of any working voltage issues.