Author Topic: Controlled Rectifiers (Read 3436 times)

SparWeb · « **on:** March 16, 2006, 06:02:04 PM »

I've discovered a variation on the typical rectifier bridge block that allows the user to control the "on-state" of the diodes. This may not be new to some members, some may in fact have 'em and take advantage of this feature, so I thought I'd post the info and see if there's any experience using these types of rectifiers.

Below is an example. There are other types.

International Rectifier THREE PHASE CONTROLLED BRIDGE

To control each Diode, you can selectively turn it on or off (one phase at a time or all at once) by applying or removing a small current to a "gate" electrode. If you use DC on the gate electrode it either stays on or off, if you use AC in phase with the supplied power then the power is rectified normally. What's really neat is that you can "tickle" it slightly out of phase with the power supply and then it will pass current only part of the time - chopping a piece off the sine wave.

I don't know of an application that does use this, but it occurred to me that this feature could be used to control the speed of a turbine, by selectively loading it and unloading it. Such as in the case of a rotor that's undersized for its alternator.

Since some members of the site like to find power in extremely low winds, this might open some doors, but I haven't completely thought that one through.

Any thoughts?

Flux · « **Reply #1 on:** March 16, 2006, 11:15:05 AM »

Yes that is a possibility, but sorting out the firing pulses for a variable frequency alternator will be fun. Also phase control has snags with harmonics in the windings and low power factor when there is much of a difference in ratio.

This is pretty ancient technology but it is much more robust than the newer devices.

I was wondering how well phase control would work a few days ago. It will certainly give you the match to the prop and the electrical efficiency will not be too bad. I don't think the axial machines will react so badly to lagging power factor as conventional slotted machines.

I have better ideas to play with but perhaps I will give it a try sometime. At least it will not be critical on layout like mosfets and Igbts.

Flux

SamoaPower · « **Reply #2 on:** March 16, 2006, 12:57:30 PM »

Flux,

Here's a paper using a phase controlled three phase SCR bridge on a wind turbine. You're right about the harmonic generation problem. I wonder about using intergal cycle zero crossing switching to mitigate this although the frequency is pretty low for it. Interesting.

http://www.itee.uq.edu.au/~aupec/aupec99/jiao99.pdf

Flux · « **Reply #3 on:** March 16, 2006, 02:33:30 PM »

Thanks for the link.

Useful comments about the unsuitability of delta in the presence of harmonics.

Flux

SparWeb · « **Reply #4 on:** March 16, 2006, 04:22:20 PM »

Thanks Samoa,

I followed the link to the various university websites and have found a lot of valuable reading material. If I could give you a "star" for that link I would. Terms like "commutation angle" and "synchronous reactance" leave me a bit bewildered, but after a bit of research I will figure out what these guys are talking about.

This paper ties in nicely with some analytical work I did recently, and once I grok this stuff I can take that work another step forward.

Thanks again.

Opera House · « **Reply #5 on:** March 16, 2006, 04:43:17 PM »

There are much better ways to do it, like pulsing a FET. Biggest problem is an extreme voltage loss of over 2 volts. That wasn't bad when they were speed controlling motors at higher voltages 20 years ago. Best way is to rectify it, use a cheap switching regulator chip and puls a FET.

Drives · « **Reply #6 on:** March 16, 2006, 06:48:15 PM »

I have to chuckle...a 3 phase SCR controlled bridge has been around since before 1960. It has been the standard workhorse as a DC motor variable speed drive, 3 phase heater control, and old AC drive "front ends". They work very well, and are extremely robust. I just worked on one today that controls the hoist that pulls salt out of a mine, 2000HP. I have been working on them for 20 years.

They would work very well as a control for a 3 phase wind turbine, I expect. The only issues I would think would be the varing frequency from the turbine, and any waveform deformation during communtation. These type of controls have to "syncronize" the timing of the SCR's firing to the zero crossing point of the sine wave. The sync circuits are designed for a specific frequency, as well as the general purpose SCR's are usually rated for no greater than 400Hz operation.

These type of controls are still produced in good amounts, the technology is well proven, however, the AC variable frequency drive is taking over fast. Many DC motors and drives are being replaced with AC induction motors, and VFDs.

Also, another 6 SCR, 3 phase coonfiguration is 2 SCR's connected in inverse parallel per phase leg. This is called a 3 phase soft start, or another form of 3 phase heater control. These would probably work better for a wind mill.

I have a few of all types laying in my "bone yard". I just need to get the time to hook them up to a variable voltage & Frequency source to test the theory.

Interesting topic!

SparWeb · « **Reply #7 on:** March 16, 2006, 11:46:14 PM »

Drives:

"...standard workhorse as a DC motor variable speed drive"

Of course! I have seen these things before, I just didn't recognize them at the time. It is much more appropriate for me to say that I'm new to electronics. A controlled SCR bridge is new to me, not the other way around.

Here's where I've been getting some of my ideas:

All About Circuits

Anyway, I'm not trying to argue a point, just drawing attention to something I think is kinda neat.

Drives · « **Reply #8 on:** March 17, 2006, 07:05:35 AM »

Steven:

Thanks a bunch for that link, I am sure it will help many people on this board.

As for the concept of using the SCR in our wind gennies, the more I think about it the better I like it. The technology is very mature, and as far as power electronics go, I have not found anything as tough as a SCR (thats thyristor for the overseas folks).

When I first found this board (read "got bit by the wind gennie bug") I tried to think of different ways to use what I know to help make gennies better.

Here are some ideas,

Use a DC drive to control current into a load (exactly what we have been talking about)
Use a Electronic soft start to control current into a 3 phase load
Use a Regenerative DC drive to generate power back onto the power line
Use a AC drive as an 3 phase inverter, hook the battery bank to the DC bus
Use a AC drive as a 3 phase heater control by feeding the gennie to direct to the bus (no need for the batts, but will need control power supply), and use the variable output to feed heater banks.

All of these options are doable, but they all need some minor "kinks" worked out. I have been hesitant to discuss any of these on here because this is a DIY board, most of these are advanced projects, and I get "fired" up when someone who does not know about the safety concerns of electrical power starts playing with this kind of stuff. (I'm going to go off on a rant for one minute, not meant at anyone particular person). I have no problem helping anyone, but when people don't crack open a book to learn about the hazards of electricity, before they ask questions on this board....it scares me to death. Ever seen a childs charred body pulled from a burned house because someone attempted to perform house wiring, and screwed up? Probably one of the worst sites you can imagine, and it could have been prevented!!! All of the above projects would probably involve 120 volts or greater....we must be careful. END OF RANT.

Now with EBAY, all of these DC drives, SCR soft starts, SCR heater controls, and AC drives are available to the public at very cheap prices.

Conclusion, if you would like to explore this some more, either email me, or leave it out on the open board if other people are interested, and we can "open up this can of worms" I have all the test equipment, datalogging, load testing, Motor Generator sets, etc you can imagine. This could be fun.

Drives · « **Reply #9 on:** March 17, 2006, 07:13:44 AM »

I left that post rather open ended...someone pick one of the above five projects, and I'll see what I can do.

(OH NO...I've committed to another project, I must be crazy!)

Flux · « **Reply #10 on:** March 17, 2006, 07:40:10 AM »

Dean

Be careful you will have me playing with this next.

The three phase converter run in inverter mode does a perfectly good job of being a grid tie inverter but problems with approval for mains. Could perfectly well be used as a powersaver with a diesel generator. This is the old Gemini inverter idea that worked well in the past.

Thyristors are the obvious choice for heating controls, the triac is a poor relation and has many problems I prefer to use back to back thyristors.

I am sure a controlled converter will make quite a good job of matching a turbine to a battery and will be much superior to a direct connection but the one thing it will not be so good at is reducing stator heating. The average ac current will fall quite nicely as the volts rise, but because of the harmonic waveform and the poor power factor there will not be such a good reduction in stator rms current, the bit that causes heating.

Even so there has to be some improvement even in this aspect.

I think results are likely to be much better with air gap alternators than with motor conversions. Small slotted iron cored machines are notorious for their leakage reactance and this is one of the biggest source of nuisance with thyristor converters.

I think you have to start again with the firing circuit for variable frequency.

Problems with device drop for the 12v boys but better at 24v and not really a problem at 48v or higher.

Flux

Drives · « **Reply #11 on:** March 17, 2006, 11:24:15 AM »

Flux:

OK, I believe we have your vote for the thyristors for heating control. :-)

Thank you for weighing in on this, I greatly respect your experience and opinion. I need some help.

For a heating type of gennie, I am guessing that we are talking a 16 foot or larger in a 9/12 configuation, wired for 48V cut in. I need to start somewhere, so correct me here if my parameters are wrong.

If we have a 48vac machine at a cut in of 7mph wind, effective running up to 28mph then I see 192vac at the top speed? 48vac is 25% of 192vac i.e. 7mph is 25% of 28 mph wind. I am guessing that we should not have more than 1KHz from the gennie?

I may be full of baloney, so I need some guidence.

I was looking at a couple of schematics at lunch, and I may be able to overcome the frequency issues. The real challenge is changing these controls to operate over a wide range of not just frequency, but also voltage. The syncronizing circuits I refer to are specifically designed to run off the stable incoming power line. 460vac, 230vac, 208vac, 60Hz, 3 phase.

Now we will be feeding both a fluxuating frequency, and voltage. To get a stable timing signal from 48-192vac is more than tricky to say the least.

WILD AC, ahh....the beauty of batteries, and resistors become crystal clear. LOL.

Thanks again.

SparWeb · « **Reply #12 on:** March 17, 2006, 12:23:40 PM »

Dean,

I didn't expect to cause so much inspiration! :^)

...stable timing signal from 48-192vac is more than tricky to say the least.

Using an air-core transformer to step down the control voltage (one on each phase, I suppose) would affect the phase angle, I guess. If you tried to correct with a capacitor, would the solution work at a wide or narrow range of frequencies? (I feel like a freshman asked to solve a second-order differential equation.)

A 16 foot prop in a 28 MPH wind, assuming a TSR of 10, is turning at 490 RPM (that estimate is probably high). A 12-pole alternator turning at 490 RPM would pulse each phase 98 times per second. Say, 100 Hz, tops.

Flux · « **Reply #13 on:** March 17, 2006, 12:48:24 PM »

To answer both here.

Yes about 100 hZ is reasonable the low speed low frequency end will be trouble.

Air cored transformers are not a reality at these frequencies and I couldn't see any advantage anyway. There is no significant phase shift with a conventional transformer.

I don't think that a commercial firing circuit is a good starting point, but I have no idea of what is used in north America.

You may be able to do limited tests with one as many machines will be near 60 Hz in the working range.

Flux

Nando · « **Reply #14 on:** March 17, 2006, 03:55:06 PM »

FLUX:

In this case, the ideal set up would be a PWM section that loads all 3 phases equally by having a 3 phase full wave rectifier.

Then depending on the power generated by the wind mill, there are two simple paths to follow.

1) The use of 3 TRIACS that load each one of the 3 phases with isolated Zero crossing Triac drivers.

If the power is much greater then one can add more than one set of TRIACS to load the generator properly.

This added load modules are set ON/OFF in sequence, like a shift register going forward or backward depending on the loading requirements, to set One, or two or three in steps.

2) The use of MosFets or IGBT loading the generator ON/OFF via the 3 phase rectifiers used by the PWM Section.

If one observes the setup, one can say that the generator is being loaded linearly from Zero to 100 %, the PWM section does linear loading and when either zero or 100 % value the circuit triggers the OFF or the ON state of a ON/OFF load.

I presented the case for a Heater control and at the same time the equivalent ELC controller is that is the case.

The circuit can have voltage and current detectors to define the load curve to follow the wind mill wind Cube power curve.

For power electric harvesting the technique in principle is about the same, a conversion module that takes the generator voltage and converts to either, charge a battery bank or to convert it to generate an AC voltage for GRID connection.

In both cases the conversion module produces the necessary voltage but the Current is varied following the Wind Cube Power Curve to insure that the wind mill does not stall and the wind mill does not over RPM.

TRIACS will operate properly to more than 1500 Hertz if the triggering uses its own incoming phase signal as explained above.

~~~~~~~~~~

Flux said:

>Now we will be feeding both a fluxuating frequency, and voltage. To get a stable timing signal from 48-192vac is more than tricky to say the least.

~~~~~~~~~

NO PROBLEM Isolation drive and proper timing allows for such wide variations in voltage and frequency.

I hope this helps

Nando

deloiter · « **Reply #15 on:** March 18, 2006, 10:38:38 AM »

Please don't try to use triacs in place of diodes (rectifiers). SCRs will work fine as a controlled bridge on a wind turbine but be careful not to assume the terms triac, thyristor and SCR are all the same. SCRs conduct in only one direction - the top or bottom half of an AC cycle - not both. Triacs normally conduct in both directions - not good for rectifying. While a properly designed triac circuit can function as an SCR, triacs are generally intended for use in gated control of current in both directions. Be sure to SCRs for gated-control rectification.

Nando · « **Reply #16 on:** March 18, 2006, 11:51:47 AM »

TRIACS AND MORE TRIACS.

Please understand the circuit WELL before you give an OPINION that may be read WRONG by many, defining the described circuit as not good.

In this circuit, I have described, the design is for TRIACS and NOT SCRs

In this circuit ( mine ) is for FULL Wave conduction.

I said ON/OFF not rectification, this to turn a Ballast ON for at least 1 Hertz and always in full 1 Hertz multiples of them, in this case the design calls for full Sine Wave ON/OFF.

jimovonz · « **Reply #17 on:** March 18, 2006, 12:45:06 PM »

"Please understand the circuit WELL before you give an OPINION that may be read WRONG by many, defining the described circuit as not good"

Good to see you took on board what I said.

Nando · « **Reply #18 on:** March 18, 2006, 02:03:33 PM »

Jimovonz:

I do not follow you here, explain and where did you commented in this thread ?.

Nando

Drives · « **Reply #19 on:** March 18, 2006, 02:14:47 PM »

OK...I don't know where I got the 1Khz from...must have temporarily lost my mind. The formula for RPM of an 3 phase induction motor is RPM=120XFreq/#of poles

If we apply this to our 12 pole gennie at 490RPM (thanks Steven) and extrapolate the formula to Freq=RPMX#of poles/120 then we get 49 Hz at top speed.

Just a point of interest at 49 hz and 192 vac, that gives us a volts to hertz ratio of 3.92 Vac to Hz. A 230 vac/60Hz motor has a V/Hz ratio of 3.83...interesting trivia.

I think you guys are absolutely correct, even with a high TSR 16 pole gennie, we would not exceed 100Hz, thanks for correcting me.

Back to our 12 pole, at cut-in speed of 25% we would be at approx. 12.25 Hz. This is not going to work for the commercial iron core transformer sync circuit. I believe we would have to use some sort of comparator op amp/opto isolator type circuit. Unless someone has another idea, this is the route I will look into. Thanks.

commanda · « **Reply #20 on:** March 18, 2006, 02:42:49 PM »

I am sure a controlled converter will make quite a good job of matching a turbine to a battery and will be much superior to a direct connection but the one thing it will not be so good at is reducing stator heating. The average ac current will fall quite nicely as the volts rise, but because of the harmonic waveform and the poor power factor there will not be such a good reduction in stator rms current, the bit that causes heating.

A 3 phase PFC followed by a pwm controlled buck converter will overcome this problem.

You might also like to read the description of the MPPT controller in my diary to tie it all together.

http://www.fieldlines.com/story/2006/3/12/14840/1315

Amanda

deloiter · « **Reply #21 on:** March 18, 2006, 02:50:54 PM »

Nando,

This thread started out talking about a "typical rectifier bridge block" and then there was a comment that "SCR is a thyristor for the overseas folks", which may be true, I'm not certain and it doesn't matter. My comments were meant to be informative in regard to the entire thread and not directed at the prior posting. My comments may not even be applicable to your circuit, I can't tell because I don't know which circuit you are referring to. I didn't mention any particular circuit and I'm sorry you thought I was dissing your design, I was not. With that in mind, I stand by every word I said, however, as suggested I will keep any future comments to myself.

Nando · « **Reply #22 on:** March 18, 2006, 04:42:45 PM »

I aggree with AMANDA:

The idea of a PFC, single or 3 phase may be a good way for power control feeding the "regulated" voltage to a DC/DC converter or a DC/AC converter.

Some GRID tied wind mills do work this way.

Though, the circuit needs MPPT detection and behavior for proper loading at maximum power level harvesting.

Generator heating reduction can be accomplished by presenting a high impedance to the generator thusly reducing the internal generator power dissipation.

Nando

SparWeb · « **Reply #23 on:** March 21, 2006, 09:35:07 PM »

I feel like such a noob, but I have to ask:

Could someone write out these acronyms, please?

PWM =

PFC =

MPPT = "maximum power point T...?"

Add any others you may feel apt to use

Redfaced,

SamoaPower · « **Reply #24 on:** March 21, 2006, 11:56:16 PM »

Don't feel abashed Steve. We who play with this stuff often tend to forget that the jargon is just so-much Greek to others. Perhaps the board could sponsor a glossary.

PWM = Pulse Width Modulation. A technique of employing a variable width pulse train to control the average voltage, current or power. Often used in switch-mode power supplies, inverters and converters.

PFC = Power Factor Control (or Converter or Correction). Many inductive type loads such as motors shift the phase of the current relative to the voltage. This can cause some problems for the supplying source. A PFC, often using a boost converter, can correct this.

MPPT = Maximum Power Point Tracker (or Tracking). A control algorithm that attempts to provide maximum power transfer between a source and a load. A form of source-load matching.

I could probably sit here all night generating a long list of others. Try using Google. Usually an encyclopedic source will pop up.

Apologies from us all who tend to forget.

commanda · « **Reply #25 on:** March 22, 2006, 12:10:02 AM »

My aplogies. That would have been me. I'll accept the slap on the wrist, and try to explain the acronyms when first used in future.

Amanda

SparWeb · « **Reply #26 on:** March 24, 2006, 01:00:58 PM »

SP:

"Perhaps the board could sponsor a glossary."

I hope you write it, then.

Thanks.

I noticed in Flux's recent (but already classic) dissertation, he mentioned rectifier control (like PWM, then), so the idea isn't new. This, like many other things I've learned from the board, will lodge itself in the back of my mind, perhaps to be drawn from the toolbox if the need arises in the future.

News:

Author Topic: Controlled Rectifiers (Read 3436 times)