Couple of nice MKIIIs

[jt1stcav]

Yeah, I know. But it still pisses me off sometimes. My dad's been working in electronics since he started repairing old tube radios at the tender age of 11 (1951)...it just came naturally to him. He tried to teach me basic electronics when I was 14 (back in '76), but it went in one ear and out the other; eventually I lost interest and so he lost patience trying to teach me. I don't have the knack to comprehend it, nor the fortitude to learn it anymore, so I leave it up to my dad and to all the other experts whenever I need assistance in anything electrical in nature. I'm just not cut out for that field of expertese...guess I get that trait from my mom (cracks my dad up whenever I tell him that)!

[sfogg]

"They work off current so with any instant on rectifier setup they

won't really kick in until the heaters of all tubes in the amp warm up."

Not even remotely. An inrush current limiter is nothing more then a NTC

thermistor. A thermistor is nothing more then a resistor thats value

changes with temperature. A negative temperature coefficent (NTC)

thermistor (inrush current limiter) is highest resistance when it is

cold. As it warms up its resistance drops.

The reason it limits the current inrush is simply because of its

resistance in the circuit... OHM's Law. It doesn't care at all what

current the amp is trying to draw at first turn on. When the amp is

first turned on the NTC thermistor is coldest.... IOW its resistance is

as high as it is ever going to be and it will performing its maximum

current limiting right then and there at first turn on.

Like any resistor as power is being disappated in it it is going to

heat up. As it heats up its resistance drops... as its resistance drops

the current limiting will be dropping too.

" I still can not believe you can't understand how a In rush current

limiter does no good if ZERO current is being drawn by a cold

compliment of tubes in the amp. They only work by the current being

drawn through them heating them up. They could careless if the

rectifier is making 500V or 50V. no current no heat"

You have this 100% backwards and misunderstand what an inrush current limiter does.

If no power is passing through the thermistor then like you said it

isn't being heated up. Which therefor means it is cold... which further

means its resistance is as high as it is going to get. It therefor is

limiting current as much as it ever is going to be.... again that pesky

OHM's Law. The more current passing through the inrush current limiter

the hotter the thermistor is going to get, the lower its resistance

which means the less current limiting it is going to be doing.

And again (going back to this same discussion from way back) the

current surge into an amplifier is *highest* when the tubes are dead

cold. Go measure the filaments on your tubes with them at room

temperature and again using that peasky OHM's Law and figure out how

much current is going to flow through them the instant you turn on the

amplifier.

For an EL-34 which has a 1.5 amp rating on its 6.3v filament the couple

I measured would allow about 9 amps *each* to pass through them at

first turn on when the filaments were cold.

I = v/R

v = 6.3v

R = about 0.7 ohms on a cold tube

I = 6.3 / .7

I = 9 amps (per tube)

After warming up you would find the filaments resistance is around 4.2

ohms if the 1.5 amp rating is accurate. The filaments are a type of

thermistor too. But they are a positive temperature coefficent

thermistor.... their resistance increases with temperature.

Shawn

[Marvel]

When your monster (or not so monster) power amplifier is switched

on, the initial current drawn from the mains is many times that even at

full power. There are two main reasons for this, as follows ...

* Transformers will

draw a very heavy current at switch on, until the magnetic flux has

stabilised. The effect is worst when power is applied as the AC voltage

passes through zero, and is minimised if power is applied at the peak

of the AC waveform. This is exactly the opposite to what you might

expect

* At power on, the filter capacitors are completely discharged, and act as a short circuit for a brief (but possibly destructive) period

These phenomena are well known to manufacturers of very high power amps

used in PA and industrial applications, but "soft start" circuits are

not commonly used in consumer equipment. Anyone who has a large power

amp - especially one that uses a toroidal transformer - will have

noticed a momentary dimming of the lights when the amp is powered up.

The current drawn is so high that other equipment is affected.

This high inrush current (as it is known) is stressful on many components in your amp, especially ...

* Fuses - these must be slow-blow, or nuisance fuse blowing will be common

* Transformer - the

massive current stresses the windings mechanically and electrically. It

is not uncommon to hear a diminishing mechanical buzz as the chassis

reacts to the magnetic stress

* Bridge rectifier -

this must handle a current way beyond the normal, because it is forced

to try to charge empty filter capacitors, which look like a short

circuit until a respectable voltage has been reached

* Capacitors - the inrush current is many times the ripple current rating of the caps, and stresses the internal electrical connections

According to this, a SS rectified amp will have more of a problem,

as they are instant on and the caps used for filtering aren't charged

at the moment of turn on. This contradicts an earlier statement. I'm

still trying to figure this out too. I pulled this off of a site that

has a design for an active ICL (using current sensing, resistors and a

relay, etc.). The single IRCL chips are way simpler to integrate into

the design, as they just go in series on one leg of your AC mains.

Bruce

[Marvel]

After warming up you would find the filaments resistance is around 4.2

ohms if the 1.5 amp rating is accurate. The filaments are a type of

thermistor too. But they are a positive temperature coefficent

thermistor.... their resistance increases with temperature.

Shawn

As is a tunsten filament in a light bulb. The resistance is very low at

turn on, and quicly goes up. If not, it would burn out very quickly.

Bruce

[sfogg]

Bruce,

"As is a tunsten filament in a light bulb. The resistance is very low at

turn on, and quicly goes up. If not, it would burn out very quickly."

Exactly, and it is during that initial turn on current surge/thermal stress when a bulb is most likely to burn out.

If the bulb is on a circuit that reduces that turn on surge (a dimmer

circuit that dims up on power ON) the bulbs typically tend to last much

longer.

Shawn

[3dzapper]

Shawn, Remember too this is the 1:20 step down side of the power transformer for the heaters. This will not effect the high tension, step up, side of the PT.

IMO adding a thermistor, which works hardest at low current/temperature, is a good means to reduce cathode stripping or arcing in the diode if nothing else.

Craig is correct too. The high voltage will build with no offsetting current until the output tubes heat enough to conduct. About twenty seconds? During most of this time, the 500V cap will be over it's rating. The builder is most likely relying on the fact that caps have an overrating cushion. This may be adequate in the short run but long term failure would be more likely.

Rick

[sfogg]

Rick,

The inrush current limiter is usually put on the AC input into the amp.

It limits the entire current the amp can take in, not just sections of

the circuit.

"Craig is correct too. The high voltage will build with no offsetting current until the output tubes heat enough to conduct. "

That isn't what Craig has been saying though with regards to the inrush

current limiter. He has said numerous times that an inrush current

limiter doesn't work until current passes through it and it heats up.

For example this:

"With a 5U4 an inrush current limiter (thermistor) will do little good

because there is no current being draw on the B+ at start up these

Thermistors work by current being drawn through them so no current

little effect. They work fine with a 5AR4 because it affords some time

for the filaments to warm up before producing B+ so when the B+ starts

rising the in-rush current limiter will kick in and slow the process

even further. Its all about experience "

Which is the exact opposite of how they actually work. The inrush

limiter is at maximum effect when the amp is first turned on... it

works with a 5U4 too. By the time the B+ is coming up on the 5AR4 the

resistance of the inrush current limiter is already dropping which

means it is letting more current through, not the other way around as

Craig described it above. On a MKIV I just tested at startup the inrush

limiter I have in the circuit is 26 ohms.. IOW limiting the amp to

about 4.6 amps of current max.

By the time you start to get any bias voltage readings (and long before

it is up to its full voltage) the inrush limiter is down to 8 ohms of

resistance which is limiting the circuit to about 15 amps max. As it

continues to heat up it drops lower still till it is just about

effectively no longer in the circuit.

Shawn

[3dzapper]

NSDT!!!

I was attempting, in a polite way, to put the proper perspective on things. A thermistor has a negative temperature co-efficient. That is, the resistance to incoming current is highest when the thermister is coolest. At normal operating temperature the thermistor injects only nominal restance.

That said, a thermister would work best against turn on surges in the power supply if the filament current were to be seperated. The high current drawn by the filaments heats the thermister too fast to be effective on the high voltage seen by the power supply capacitors. Thus there would be no effective protection for that 500V capacitor.

Rick

[sfogg]

Rick,

"The high current drawn by the filaments heats the thermister too fast

to be effective on the high voltage seen by the power supply

capacitors. Thus there would be no effective protection for that 500V

capacitor."

OK, gotcha... wasn't sure what you meant, thanks for the clarification.

WRT to overvoltage on the PS caps in a MKIII I'd agree, the inrush

current limiter likely isn't doing much of anything for that with or

without the 5AR4. If anything it might be making it slightly worse

since the heating of the tubes will be slightly delayed by the current

limiting of the thermistor and therefor it will take them slightly

longer to start conducting. That might even out with a 5AR4 since it

will be slightly delayed in heating too.

Shawn

[3dzapper]

Also, although most thermistors are placed in the AC line-in, with the proper voltage rating, there is nothing wrong with inserting one on each leg of the rectifier cathode circuit. My ST-35, designed by Shannon Sparks, uses one in the high tension circuit just prior to the SS diodes.

Rick

[Erik Mandaville]

Despite the pros and cons of this issue, my point in bringing this topic up had to do with what IME is an outstanding pair of amplifiers, very well built and finished, for what is also IME and MO a very reasonable price. This individual is retired, and works almost constantly on this stuff. He has gotten orders from all over the world, and has a good reputation for both quality and customer support.

Considering the cost, replacing a filter capacitor, if needed (I've seen even brand new capacitors with very safe margins in working voltages) is still very small in terms of what one has already saved with a high quality pair of amps such as these -- or the Dyna 70s he does.

Thanks for the clarification on the thermistor in relation to the rectifiers in question. I just didn't have it in me to go through it all last night.

Erik

[Marvel]

Also, although most thermistors are placed in

the AC line-in, with the proper voltage rating, there is nothing wrong

with inserting one on each leg of the rectifier cathode circuit. My

ST-35, designed by Shannon Sparks, uses one in the high tension

circuit just prior to the SS diodes.

Rick

Rick,

The thermistor in Shannon's ST-35 is on the 115 volt side of the power

supply tranny, so it works on the whole circuit. That was the point of

my quote from Rod Elliott in Australia. Even the ps tranny will get a

spike at turn on because of the magnetic field being built in the

coils. Very quickly, but it is there. Plus what hits the diodes can be

far beyond the rating of the diodes.

This contradicts Craig's earlier post. I'm NOT knockin' his work, it is

exceptional, but the current limiter is CHEAP, and helps protect more

that the ps caps.

Perhaps, instead of the standby switch with the resistor we were

talking about earlier on the center tap of the high voltage output you

could just use an IRCL there. Then you would gain the benefit of a

'soft start' on the output tubes. I'll email Shannon and ask him about

the idea.

Bruce

[sfogg]

Bruce,

"Perhaps, instead of the standby switch with the resistor we were talking about earlier on the center tap of the high voltage output you could just use an IRCL there. Then you would gain the benefit of a 'soft start' on the output tubes. I'll email Shannon and ask him about the idea."

One inline with the high voltage would limit current but I don't think it would help with the over-voltage.

If you can find an appropriate one you *might* be able to use a positive temperature coefficent thermistor (resistance increases with temperature) between the high voltage and ground to dump some of the excessive voltage on turn on when the tubes aren't yet conducting. You would want one with a high initial resistance though to limit current flow through it to ground to prevent damage to the HVPS. And it would need to be fairly quick acting with regards to temperature (resistance increases quickly with small temperature changes) to get high enough in resistance to effectively get out of the circuit by the time the tubes are warmed up enough to start conducting. There might not be enough current in the high voltage to heat up the PTC thermistor enought to increase its resistance though.

If that was the case a time delayed relay could open/close the HV or open close a voltage divider to cut down the level for a short amount of time till to lower the voltage for a bit to allow the tubes to start conducting.

Shawn

[NOSValves]

Shawn,

Do me a favor hook and 100 Ohm resistor to a power source with no current drawn through it! The voltage will be Identical on both sides of the resistor. The same is true of the inrush current limiter or any type resistor. I am not reading this from a book I'm taking from experience. When you fire up a Mark III with a in-rush current limiter/5U4 the voltage will spike above the cans rating until the tubes warm up. If the 5AR4 is used with the Thermistor the voltage is delayed by the 5AR4 giving the tubes a chance to warm up when the 5AR4 starts making B+ the tubes are ready to pull some current so the voltage spike is slowed.

I suggest you quit using techno babble and take a tube rectified tube amp with a 5U4 or a SS rectifier for that matter and install a inrush current limiter in it and see for yourself. Yes for a very brief instant the thermistor will see some current but that stablelizes in a split seconds not even long enough for a meter to register it. after that no major current is drawn through the thermistor until the tube heaters warm up the cathodes about a minute so the resistance does no good without major current there is no sunstancial voltage drop across the thermistor just like a resistor.

This is the constant battle here between the technically minded and the technically experienced. How things are drawn out on paper is not always how they work.

Yes I could of explained the thermistors relation to heat better but I was sure you would come to your little buddies aid and make a techno babble attempt at saving him anyway.

The only way a thermistor will work to stop in-rush current with a instant on B+ voltage source is with a stand by switch to get the heaters warm on the tubes. Been there tried it, just installing one does little or no good at all.

Craig

[3dzapper]

"The thermistor in Shannon's ST-35 is on the 115 volt side of the power supply tranny, so it works on the whole circuit."

Correct, I mis-spoke. My bad.[:$]

"This contradicts Craig's earlier post. I'm NOT knockin' his work, it is exceptional, but the current limiter is CHEAP, and helps protect more that the ps caps."

Don't knock Craig on this one. He does use IRCs on his VRDs and on rebuilds. This even though he has a bunch of overvoltage reserve built into his PS caps. Most likely the voltage there is limited only by the bypass cap if it/they are 630V rated.

On my 300B amp, used 100 Ohm 10W resisters on the cathode legs as secified in many designs. Again this lets me "get away with" increased PS capacitance without arcing the 5U4.

Erik, The rebuilder makes a goodlooking amp but, $1100 is on the high side for Mark IIIs with original circuitry and reduced PS cap protection. He doesn't mention it but, perhaps, he has an IRC in the circuit. For not a lot of money, using readily available surplus capacitors he could easily have installed better surge protected caps under chassis. Using the above chassis cap installed as a 25uF/1000V series pair for the input.

If I ever get done with my Mark III amps, I will be into them for not much less than he is asking plus a LOT of time and agrevation.

Rick

[NOSValves]

Rick,

It does little or no good where the inrush current limiter are placed in the circuit, without current they do nothing just like a resistor. Pull all your tubes out of a amp and turn it on briefly testing the voltage in the power supply the voltage will be the same from the first stage to the last. No current no voltage drop. This is why a good designer uses near or the same rating power supply cans from start to finish in there amp because everytime you turn on the amp the entire power supply is subject to the same high voltage as the first stage. Same for coupling caps!

Techo bableling back and forth doesn't change what works and doesn't work.

Shawn.

I stated right from the beginning that the In-rush current limiter works off current which is 100% correct. I may have not explained what it does when it heats up properly but trust me Shawn your not telling my anything I don't already know. I just have poor writing skills plain and simple. As usual your running around trying to nit pick what I say apart. I for the life of me can't figure how you can not understand why this pisses me off.

Craig

[3dzapper]

Craig, Like in sex, the first 3 seconds are important. Yes the caps rating in the Mark IIIs will be exceeded until current flows in the HV, just not "slammed" as hard. The longer RC time constant will allow a little ramp up anyways to save some of the stress on the tranny, choke and rectifier cathode on initial turn on, momentarily the highest current the amp will see. For under a dollar is there any reason to omit one?

Rick

[NOSValves]

Also, although most thermistors are placed in the AC line-in, with the proper voltage rating, there is nothing wrong with inserting one on each leg of the rectifier cathode circuit. My ST-35, designed by Shannon Sparks, uses one in the high tension circuit just prior to the SS diodes.

Rick

Rick,

The thermistor in Shannon's ST-35 is on the 115 volt side of the power supply tranny, so it works on the whole circuit. That was the point of my quote from Rod Elliott in Australia. Even the ps tranny will get a spike at turn on because of the magnetic field being built in the coils. Very quickly, but it is there. Plus what hits the diodes can be far beyond the rating of the diodes.

This contradicts Craig's earlier post. I'm NOT knockin' his work, it is exceptional, but the current limiter is CHEAP, and helps protect more that the ps caps.

Perhaps, instead of the standby switch with the resistor we were talking about earlier on the center tap of the high voltage output you could just use an IRCL there. Then you would gain the benefit of a 'soft start' on the output tubes. I'll email Shannon and ask him about the idea.

Bruce

Marvel,

Lets clear something up I'm not saying don't use in-rush current limiters in anyway. I use them in my Mark III redesign on every set of amps I do this with a 5AR4 they work perfectly. I'm saying with this particular amp with the WRONG rectifier and under rated power supply cans it simply will not work in the LONG HAUL. No more no less.

Someone trying to say that this seller has some long term track record with this setup that can be drawn from for some kind information pool is full of it. Where was he a few years ago? Heck even a year ago. I'm sure he is a nice guy he was nice to me in corrospondence but I'm sorry he is cutting corners where they just don't need to be cut plain and simple this is my entire point! As is... the way he is building the amps that 500V can has no business being there.

This was the main point of my post and it has been side stepped into a spectulative cure that is a simple fix ............USE the proper rated parts and there is no problem!! But hey this would cost about $40 more.

Craig

[NOSValves]

Craig, Like in sex, the first 3 seconds are important. Yes the caps rating in the Mark IIIs will be exceeded until current flows in the HV, just not "slammed" as hard. The longer RC time constant will allow a little ramp up anyways to save some of the stress on the tranny, choke and rectifier cathode on initial turn on, momentarily the highest current the amp will see. For under a dollar is there any reason to omit one?

Rick

Rick its like a bandaid the first 3 second are not going to save the can in the LONG HAUL. Why is this so hard to understand??? Use the right parts with a 5AR4 and in-rush curent limiter on the AC input and you have a nice behaving B+! Trust me you would be amazed there IS NO VOLTAGE SPIKE AT ALL !!! As the 5AR4 starts making B+ it comes up to about 400V and then slowly rises to about 480V with a wall voltage of 117VAC. With a 5U4 it sky rockets above the can and sits there for near a minute with or without a inrush current limiter. No theory here been there, done it, tested, end of discussion. No need for techno babble.

Craig

[NOSValves]

NSDT!!!

I was attempting, in a polite way, to put the proper perspective on things. A thermistor has a negative temperature co-efficient. That is, the resistance to incoming current is highest when the thermister is coolest. At normal operating temperature the thermistor injects only nominal restance.

That said, a thermister would work best against turn on surges in the power supply if the filament current were to be seperated. The high current drawn by the filaments heats the thermister too fast to be effective on the high voltage seen by the power supply capacitors. Thus there would be no effective protection for that 500V capacitor.

Rick

Rick,

I missed this post!! By George some one has got the point.

Craig