How to test a Bi-Directional Triode Thyristor

[jim hardy]

meBigGuy is thinking straight.

I too would like to see 'scope traces at top of R403-C405 to see of the cap is getting charged
and gate of triac to see if any of the charge on C405 makes it on over to gate

 

[Salvador]

I wanted to say that actually I'm quite happy that I jumped into this thread, at first I was much like big guy already said kind of stupid towards the whole concept , went on with what I think it does instead of looking at what it really does.
I must admit I'm taken by surprise.
But happily I now know better.I must say of all the HI FI bla bla bla stuff which is there like 80% of the time and except the sales things of the Carver , this power supply idea is rather simple but very cool , he really manages to squeeze a mains transformer to it's max with not that many extra parts.Quite interesting.
Basically Jim's post about the tops of the rectifier cycle and another place at the internet gave me some clues as one other folk had mentioned about the ignition coil thing besides big guy at one other place.

 

[jim hardy]

as one other folk had mentioned about the ignition coil thing besides big guy at one other place.

i'm still chewing on that one



Thyristor stops conducting when current through it is zero.
At the instant of zero current there should be zero energy in the transformer, energy being ½LI² with I=0 .
So which is true - intuition or math ?
Actually the the thyristor ceases conducting when current through it drops to its "holding current" value , near zero but not quite.
from its datasheet at http://www.datasheet.hk/view_online.php?id=1927120&file=0465\sm16gz41_6629879.pdf

So i am relieved - the "ignition coil" effect is small but very real and there. Math and intuition agree.

meBigGuy is right the filter caps enable graceful turnoff by capturing the stored energy.
R310-C404 help too.

Hang in there Salvador - we all continually add to our understanding one tiny piece at a time. I had missed the significance of C405.

my two cents.

old jim.

 

[Salvador]

well correct me if I'm wrong but I guess I'm starting to get there myself , the whole " secret" is in how and when you turn on or off the sine wave.
if I'm correct the magnetic field strength is larger when the rise or fall of a waveform is steeper or closer to vertical and also when the point at either turn on or off is at a higher potential , in other words if you switch the sine wave to the transformer primary while it's at about few volts above zero the field begins weak and then builds up with the sine wave but that happens evenly and quite slowly the same with turn off , but if you turn on the TRIAC when the sine wave is already at say near maximum , it has high potential at that point so the primary current starts quite quickly and the rise is fast so the field in the transformer starts much stronger than when it would begin together with the sine wave.

this is why they call it an " ignition coil" and ignition coil wo5rks much like an smps only the switching is the other way around , but in both cases you rapidly and suddenly apply a full steady DC potential to where no current and voltage was before and that gives you a rapid change in flux and if the energy source is strong enough a lot of power with it.
the sine wave starts much slower and ends the same way so it doesn't create the fast change in flux and at lower than peak voltages were the caps are already full it simply goes in and back away without much use except for winding loss and stray fields.

this is how I have came to see this if there is something wrong I would appreciate if someone corrected me , thanks.

 

[Planobilly]

As soon as the parts arrive and get installed I will make the measurements you guy ask for. I have replaced or tested everything from the mains to the transformer. The only exception are the two diodes D402 and D403 which were 1n4002 which got replaced with 1n4004's. 1n4002's are on order. The fuse in the transformer is working because voltage got to the three main bridge rectifiers when I jumped the triac.

Edit: also I do not understand the data sheet regarding the triac. What is the max voltage that can be applied at the gate?

 

[jim hardy]

Edit: also I do not understand the data sheet regarding the triac. What is the max voltage that can be applied at the gate?

It's sort of like a zener diode , at some unspecified forward voltage it'll start accepting gate current
They're telling you that unspecified forward voltage will be not more than 3 volts
http://www.datasheet.hk/view_online.php?id=1927120&file=0465\sm16gz41_6629879.pdf

and triac will "fire", ie commence conduction, at gate current ≤ 50 milliamps.
A line or two later they tell you the unspecified gate voltage will be not less than 0.2 volts, so we should see gate pulses between 0.2 and 3 volt .
Up in Maximum Ratings they tell you that if you want to use the triac ever again, don't force enough current into the gate to exceed ten volts between gate and T1.

 

[jim hardy]

if I'm correct the magnetic field strength is larger when the rise or fall of a waveform is steeper or closer to vertical and also when the point at either turn on or off is at a higher potential , in other words if you switch the sine wave to the transformer primary while it's at about few volts above zero the field begins weak and then builds up with the sine wave but that happens evenly and quite slowly the same with turn off , but if you turn on the TRIAC when the sine wave is already at say near maximum , it has high potential at that point so the primary current starts quite quickly and the rise is fast so the field in the transformer starts much stronger than when it would begin together with the sine wave.

There's a long long thread "how power transfers across an ideal transformer"...

Go back to your basics
B=μNI/L_{ength of magnetic path}
The only term that's a function of time is current I .
So magnetic flux is in proportion to current.
The derivative relation between flux and voltage makes it counter-intuitive until you have accustomed your brain to work that differential .

So break up your run-on sentence into bite sized thoughts,
one step at a time,
one thought per line
and your thinking will get better.

AC circuit theory is usually taught without emphasizing that sinewaves are a mathematical oddity -
they and their derivatives look just alike.

 

[jim hardy]

The fuse in the transformer is working because voltage got to the three main bridge rectifiers when I jumped the triac.

That fuse is for gate drive current, not power

with it unplugged - got no ohms between p5 and p6 ?

on another track.......

The PCB track where D402's anode lands - any components nearby where a previous troubleshooter might have accidentally shorted something?

Seems to me this area would require caution when poking around.. check for smoke residue

 

[Planobilly]

Hi Jim,

I built this circuit and why I was asking about the max gate voltage. The lamp was 72 watts, as I did not have a 100 watt bulb. Close SW1 and the lamp lights as would be expected.

I also replaced the 100 ohm resistor with a 1000 ohm resistor. The voltage was 115 VAC 60hz. I did not have a heat sink attached to the triac so I did not leave it on for too long.

So, with this circuit, based on a 72 watt lamp ( lamp resistance of 15 ohms apx. Z=?) and a 1000 ohm resistor and 115 VAC 60hz what is the gate voltage?
I am still stuck on stupid trying to understand the triac...lol

To answer your question about D402 and the area near it...no sign of any shorts. I replaced all the resistors, the two diodes, and the four transistors connected to D402. I have tested the bridge rectifier and the optocoupler which I think are OK.

Cheers,

Billy

Update: after replacing the filter cap I smoked and disconnecting the 100V secondary the triac is working and sending current to the transformer. That is the good news.

Now I need to find the fault in the 100V part of the amp and when I do I think everything will go back to normal. This amp is a PITB to work on...lol...Don't try this at home...lol

 

[jim hardy]

what is the gate voltage?
I am still stuck on stupid trying to understand the triac...lol

The gate voltage in your TEST circuit will be a squarish-looking wave at line frequency
with flat tops between 0.2 and 3 volts
and maybe a spike at front edge that reaches 3 volts.
So a DMM set to AC should read between ~0.15 and ~2 volts.

The gate voltage in your amplifier will be a short pulse once per line half-cycle, again between 0.2 and 3 volts.
It's narrow because its source is C405.
So it has not enough area to cause a DMM to show much of anything. You'll have to measure it with a 'scope.

That's my thinking.Think of the Triac as a snap action switch , initially open,
that snaps closed when gate current reaches a few milliamps
at which point all 3 terminals become very nearly shorted to one another
and remain so until current falls to zero with next sinewave zero crossing , at which time it snaps back open.

That's your simple mental model.

Your "new" SCR manual describes in detail how the gate strikes conduction in a small region and it spreads out from there as current rises , That's why they need di/dt protection. And why it has a finite turn-off time.

A useful device to have in your "bag of tricks" - can be a power control device, also a memory device.

old jim

 

[Planobilly]

OK...Thanks Jim, I think I understand now. Looking forward to reading the SCR manual. Books are generally slow to arrive so it may be a bit before it gets here.

As a side bar I was able to sell two of the Fender Hot Rod DeVille amps so I got more money to feed my amp addiction...lol All these electronic toys seem to be multiplying like rabbits. I need a smaller house and a bigger shop but don't let my wife know I said that...lol

It's back to the farm early in the morning. I have work to do on the back hoe and the potting machine. It helps if one is not put off by the smell of cow manure and diesel fuel !...lol

Cheers,

Billy

 

[jim hardy]

I have work to do on the back hoe and the potting machine.

agreed on shop ...

I have to put torque converter back in neighbor's backhoe. Never a shortage of things to fix, is there ?
Ah well this is after all the age of technology and it should be enjoyed..

IMHO high school should teach more physics and appliance repair .

 

[Salvador]

hopefully we will one day get to the age were technology repairs itself much like robots build cars in an automated conveyor line these days. And then we will be finally able to proceed to the next step which would either be finding out aliens and fighting with them (Star Wars) or the rebellion of the machine and civilization would be taken over by robots (I like Stanley Kubrick's Space Odyssey)

as for the thread topic , I didn't understand from your last response to my post Jim , so was I right in the understanding of this psu design or not or half way there ?
Now it seems logical that opening a primary loop while the sine wave through it is still near its peak causes a larger B field to develop in the core than if the sine wave goes down to zero the usual way it does.Although I guess one would have to calculate the field strength in the switch opening case VS the one you would get if the current runs the way it normally does.
Although Carver talks about getting as much out of the mains through a given size xformer as he can so since the secondary has cap bank to feed this sudden current on/off switching may do the trick.

 

[jim hardy]

so was I right in the understanding of this psu design or not or half way there ?

I apologize Salvador... Iwas having so much trouble parsing out that long long long sentence into individual thought steps that i gave up.

if I'm correct the magnetic field strength is larger when the rise or fall of a waveform is steeper or closer to vertical and also when the point at either turn on or off is at a higher potential , in other words if you switch the sine wave to the transformer primary while it's at about few volts above zero the field begins weak and then builds up with the sine wave but that happens evenly and quite slowly the same with turn off , but if you turn on the TRIAC when the sine wave is already at say near maximum , it has high potential at that point so the primary current starts quite quickly and the rise is fast so the field in the transformer starts much stronger than when it would begin together with the sine wave.

that doesn't sound right to me.
One step at a time. Transformers first.

Here's steady state flux and voltage.

if you energize a transformer by closing the switch(or Triac) at the voltage peak , you start with flux at zero and voltage at peak, which is normal and it's a gentle start for the transformer. That's why they sell solid state relays that switch at peak instead of zero crossing for use with inductive loads.

If you energize a transformer at the voltage zero crossing, you stare art with flux and voltage both at zero which is not normal and some period of adjustment will be required for the flux-voltage relation to establish normalcy.

from http://www.allaboutcircuits.com/textbook/alternating-current/chpt-9/practical-considerations-transformers/

since the transformer core will probably saturate there'll be a current inrush

That inrush can persist for several cycles
from a decent article here http://www.globalspec.com/reference/74817/203279/10-3-transformer-protection

you need to think derivative. For that it's helpful to use non-sine wave.

so closing the triac near peak is easier on the transformer's iron.
But it gives a current surge into the filter capacitors.

You might peruse that ideal transformer thread.

 

[Salvador]

Hmm I will have to read up some more on this but it sounds weird the fact that at zero crossing or starting from zero the flux is higher than when the primary is connected at near peak of half wave.
I was thinking like this , when you are near or at zero the AC value is basically zero for a very short moment of time there is no voltage , that I think we can agree upon.
This is what I can't understand , how can it be when you basically connect almost no voltage to an inductive load there develops a bigger than usual flux and inrush current respectively ?
I've always though it's about how fast the change happens from zero to max that determines many things including core saturation inrush current and flux.
in an smps for example each time a semiconductor connects the primary to the DC load the primary starts off from usually no voltage to suddenly up to max supply voltage and this creates a large flux in the core.
if each of the cycles would be made longer than necessary the core would probably saturate and the current inrush would get even bigger but I can't see how can one saturate a mains iron core with a since wave starting at zero.

Ok I will go read some on my own too otherwise I am turning this into a theory discussion

 

[dlgoff]

I need a smaller house and a bigger shop ...

My whole house is becoming a shop.

 

[meBigGuy]

At any point in time the triac can turn on and the transformer field (current) builds up. When the triac opens, the field collapses causing a voltage surge in the secondary which is captured by the filter capacitors.

 

[jim hardy]

..

When the triac opens, the field collapses causing a voltage surge in the secondary which is captured by the filter capacitors.

That's how a flyback converter works

but to pick nits, the triac won't open until current through it falls to zero (well actually to less than its holding current, 50ma for this one) .

caution - that's the current zero crossing which is not exactly at the voltage zero crossing.

 

[Planobilly]

If all this, let's call it "Carver Power Amp idea" is really such a good idea, could this type of power supply be used in guitar amps? If so, why is it not being done?
I have never met a guitar player who is happy about how heavy his amp is. This is really true with 100 watt plus guitar amps. Ever load a 100 Watt Fender Twin Reverb into the club? Is there anything special about tube amps that would make this system not workable?

Cheers,

Billy

 

[NascentOxygen]

When the triac switches off, it opens the primary (like an ignition coil) and causes most of the transformers magnetic field energy to dump into the output capacitors.

I don't think so!

 

[meBigGuy]

I don't think so!

I'm glad for you, but that's the way it works. Did you read the white paper? I know it's a bit short on hard technical details (to say the least), but that's essentially what it is saying.

BTW, the waveform in Jim's post is Voltage. A standard transformer conducts into the filter caps at the voltage peaks.

I don't know how he is shutting it off at peak current so it can dump into the caps. Let's focus on that for a bit. After-all, such a thing is certainly possible. That is, the " field collapse dump" idea can work, so why wouldn't the circuit do what he says it does.

 

[Salvador]

I think Planobilly the tube amps don't have much use for this type of psu because their load is rather different from a semiconductor amps load. most of the load in tube amps is actually constant because it goes to the heating of the filament which is constant and rather huge especially for big tubes.
I don't know specific numbers but I assume the amps musical output takes up to no more than 1/3 of the other " waste" loads that are supplied all the time , most of which goes to heating filaments then some goes to cathode idle current etc.
for a tube amp I think a well made very "silent" smps would be the weight reducer but you would still need an iron output transformer so half the weight still would be there.
unless someone maybe has an idea about a class D tube amp :d I don't know if that's even possible.But surely would be a bad idea.

 

[Svein]

unless someone maybe has an idea about a class D tube amp

Using tubes as switching devices was tried in the earliest computers. It was not a success.

 

[Planobilly]

Hi Salvador,
I guess guitars players will be out of luck for a while. Perhaps someone in the future will find a way to make light weight transistor guitar amps that will meet the needs of guitar players. For bass guitar, transistor amps are very common and Harke is now making amps with switching power supplies.

Back to the Carver. There is a long conversation at http://www.diyaudio.com/forums/everything-else/69614-carver-magnetic-field-power-amp-whazzit.html
It was long and not easy to read but it did give me a better idea about Carver amps in general. The main guy that was posting said his shop was the factory Canadian warranty station and seemed to be up to speed on Carver. He at one point or the other described the issues I am having dealing with the repair of this amp.

Based on what I have learned so far, I will muddle through this and fix the bloody thing but I don't think I will be in a big hurry to to work on any other Carver amps.
People who are liking this kind of amp also seem to be liking electrostatic speakers which I would not want for free. Each to his own I guess. Of all the Carver amps the M1.ot I have is not well liked as Carvers go. For sure this will be a "fix it and sell it" project.

The biggest issue with working on this amp is how to find the fault It looks like there is no easy answer to that question. To add insult to injury, very small issues with the amp can cause big damage. It appears that one needs to be very up to speed on this amp to work on it. Way passed my skill set at this time but not the first project I have taken on that was over my head! Most likely will not be the last..lol

Thanks for all the education you guys are giving me.

Cheers,

Billy

 

[jim hardy]

I don't know how he is shutting it off at peak current so it can dump into the caps. Let's focus on that for a bit. After-all, such a thing is certainly possible. That is, the " field collapse dump" idea can work, so why wouldn't the circuit do what he says it does.

Had Mr Carver's figure looked like this

i would agree with his explanation.

As to inductive "kick"
The triac will turn off only when current through it has decreased to less than its "holding current " of a few tens of millimaps.
So i expect a small but finite "kick" effect. Perhaps it's enough to power the amp at comfortable listening level of a watt or two..
120 volts X 50ma is after all 6 watts...

Mosfet SMPS's do switch off at substantial current but i think Carver's power supply is more akin to a lamp dimmer .

old jim

 

[jim hardy]

@Planobilly

how's it going?

I didn't get very far in that long conversation.

Where are you at with the amp now ?

Have you figured out how to check the transformer ? Easy if you can disconnect the secondaries...

old jim

 

[Planobilly]

Hi Jim,

The transformer is ok. The fault is in the 100V part of the amp. I disconnected the 100V secondary from the transformer and the triac is working now. I am still looking for parts to show up. Perhaps this week.

Knowing what I know now I understand a little better how to go about troubleshooting this amp. There is no easy way! Check every transistor. Check every opamp. Check every diode, Replace all the electrolytic caps as there are all old. Jeezzs...Oh well, such is life.

At least not having the parts has given me the time to finish a few other projects. I have a Gibson ES 335 that I have been putting off replacing tone caps on. I got it done but it is like building a ship in a bottle! Much teeth nashing and a few rude words in Italian...lol

Cheers,
Billy

 

[jim hardy]

The transformer is ok. The fault is in the 100V part of the amp. I disconnected the 100V secondary from the transformer and the triac is working now.

What Good News !If you're able to get full line voltage across transformer primary, 120V,
that puts to rest one of the (unlikely) suggestions from that diyaudioblog , namely that the transformer is some kind of saturable reactor- because it doesn't saturate and there's not a DC control winding on it .

This part of 100V supply is easy 'nuff

any idea how 100V gets to those opamps that control the photodiode?
If + and - 100 are unbalanced they'll do something to the bottom opamp via R308(or is it 306?) and 309...

i don't have a grip on it yet.
Remember "Deer Hunter" ? It's challenge - nail that trouble in one shot.

I do not understand the significance of that heavy black line that goes so many places.

None of my business i know, but it's sure an interesting circuit.

old jim

 

[dlgoff]

There is no easy way! Check every transistor. Check every opamp. Check every diode, Replace all the electrolytic caps ...

Welcome to the shotgun approach to troubleshooting.

 

[Tom.G]

I do not understand the significance of that heavy black line that goes so many places.

It is circuit common, or Ground in the vernacular. (notice it connects to the low side of the audio input connectors in the upper left of the schematic and the CT of the pwr xformer secondary.)

 

[jim hardy]

Thanks Tom -
i may have missed a tie point. That makes sense now.

 

[jim hardy]

While 100V secondary lifted: 100 volt filter caps both good ? 100V Bridge ? Winding not grounded ?

 

[Planobilly]

Hi Jim,
In reading the many things that have been said both here and on other forums, it is clear that there are many opinions as to how this amp works. I am most likely the least qualified to make any determination on my own. I also do not understand the heavy black lines. It is also hard to read this schematic for many reasons. The symbol for the triac did not look like anything I had seen before, nor have I find found one just like it elsewhere. Perhaps it is normal and just not known to me.

I will have to do some digging to find out where and how the 100V supply gets to things.I have all the electrolytics removed from the amp at the moment. When I get the parts and things back to normal I will continue the troubleshooting.

As far as the transformer goes there are only three wires going in and six wires coming out. Two each on the secondary for the three voltages and two wires on the primary from the mains, one going through the triac. Also one wire from the thermal safety device. I have not removed the transformer so I guess something could be in a place I can not see but I don't think so. I do not think there is anything special about this transformer as far as I can see at the moment. Someone would need to put me up to speed on how DC could be used to control a transformer in the first place. At the end of the day, three AC voltages come out of the secondary 30V, 60V, and 100V and go to the three bridge rectifiers providing positive and negative supply voltages.

I got the idea from some post that the triac (or what ever was controlling the triac) was not only controlling the timing of the AC but was controlling the voltage to the transformer. I have no idea how that could be possible. Perhaps when I get the book you recommended I will have a better idea of what a triac can be used for. Where ever I got the idea I remember that person saying the voltage to the primary was about 40V. I did not measure the voltage to the primary. I measured the voltage on the secondary after I removed the 100V connection from the transformer to the rectifier. At that time I was not comfortable letting things being powered up and only made a few quick measurements.

And Don...I had given the shotgun approach some thought...but more along the lines of using my 12 gauge on the amp...lol

Cheers,

Billy

 

[Planobilly]

Thanks Tom...good eye!

 

[Planobilly]

While 100V secondary lifted: 100 volt filter caps both good ? 100V Bridge ? Winding not grounded ?

I removed and am replacing all the filter caps Jim. Also 100V windings are good.