Using SCR's as a rectifier bridge (experimental)

[JW]

Using SCR's as a rectifier bridge

Hi All,

This a trackback to another thread and started here-

http://www.fieldlines.com/index.php/topic,146527.msg1003584.html#msg1003584
This is a trackback for this thread-

I had a couple more of those SCR's that I dug out of my collection of spare parts.



http://www.allaboutcircuits.com/vol_3/chpt_7/5.html

http://www.fieldlines.com/index.php/topic,146527.msg1003584.html#msg1003584

JW
« Reply #15 on: March 15, 2012, 06:55:20 PM »
I would use a 50amp breaker (self-reset) for the DC lines... I have also found you can use a 300amp SCR as a rectifier in a bridge configuration (6/w 3phase) for cheaper than the rectifier diode listed, and the thread dia is about a 1/2in to the heatsink, and the trick is to leave the gate termial isolated, in other words you can get a SCR with higher ampere rating for less the cost of a large package rectifier diode...

http://www.allaboutcircuits.com/vol_3/chpt_7/5.html
Image's from link


*
bvan1941
« Reply #16 on: March 15, 2012, 07:01:47 PM »
jw,
great observation and recommendation on the SCR's !!!
It's usually not thought of that way. Glad to see you point that out
Re: Light Weight Wind Turbine Project - as I build it.
« Reply #17 on: March 15, 2012, 07:10:49 PM »

I have also found you can use a 300amp SCR as a rectifier in a bridge configuration (6/w 3phase) for cheaper than the rectifier diode listed, and the thread dia is about a 1/2in to the heatsink, and the trick is to leave the gate termial isolated, in other words you can get a SCR with higher ampere rating for less the cost of a large package rectifier diode...

*
dinges
« Reply #17 on: March 15, 2012, 07:10:49 PM »

Are you sure about using SCRs as rectifiers? Don't think it works like that. Just grabbed a few large SCRs out of the box and none behave as a diode, with the gate left unconnected.

The link (allaboutcircuits) shows that when the gate is left disconnected it behaves as a Shockley diode (which is a 4-layer diode). Note that this is not a Schottky diode! (which is a 2-layer diode).

If I'm wrong, please correct me
*
JW
« Reply #18 on: March 15, 2012, 07:34:29 PM »

Are you sure about using SCRs as rectifiers? Don't think it works like that. Just grabbed a few large SCRs out of the box and none behave as a diode, with the gate left unconnected.

Well this one does, as you can see from the picture I have my Fluke 88 connected in "diode test mode" with the gate isolated and its passing the diode check...

Are you sure to have polarity correct? Im getting a reading of less than one ohm.
*
dinges
« Reply #19 on: March 15, 2012, 07:54:58 PM »

Are you sure to have polarity correct? Im getting a reading of less than one ohm.

Yes, I see. It appears the SCR is defect. Have you tried measuring it with other polarity? (reversing the meter leads)

Notice that in the allaboutcircuits article that in the schematic, they talk of controlled rectification; they've left out the electronics to drive the gates of the SCRs, but that's definitely needed to use SCRs as rectifiers (as e.g. in synchronous rectifiers).

Tested a few more known-good SCRs, and all measure fine on the semi-conductor analyzer (with ability to measure the most important parameters of SCRs), but none show any diode action between K and A; which is to be expected, because in that case the Gate would be useless....

My suspicion is that your SCR is defect and is a dead short. By reversing the DMM leads you should be able to verify for sure.

Either that, or my knowledge about SCRs is seriously lacking....
*
JW
« Reply #20 on: March 15, 2012, 08:08:45 PM »

My suspicion is that your SCR is defect and is a dead short. By reversing the DMM leads you should be able to verify for sure.

I just did that, and you are correct its a dead short. interesting.

Just as an aside I abandoned any attempts to use SCR's with the switching arrays im running now, Mosfets when syncronised can give those low impeadance (switching) readings using DC. The problem with SCR's is that they are subject to 0volts threshold, so SCR's work great as diodes when used in the bridge configuration. The schematic's are pretty clear as to DC output. So for the same reason an SCR is no-good for a DC switching application (mosfets n/p channel) it is usable as a rectifier when the gate is not energized.

they talk of controlled rectification; they've left out the electronics to drive the gates of the SCRs, but that's definitely needed to use SCRs as rectifiers (as e.g. in synchronous rectifiers).

Im going to look for where it says the gates need to be rectified to convert AC to DC
*
JW
« Reply #21 on: March 15, 2012, 08:14:10 PM »

aH Ha now I see what you mean
*
JW
« Reply #22 on: March 15, 2012, 08:27:05 PM »
Im just wondering, is it possible to cast a trigger coil in the stator, if anyone wanted to take the leap too-

but that's definitely needed to use SCRs as rectifiers (as e.g. in synchronous rectifiers).

they've left out the electronics to drive the gates of the SCRs,

"synchronous rectification" SCR's are way cheaper than rectifier diodes for current capacity, so if you could make a cheaper rectifier with a simple trigger circuit (cast into the stator) wouldnt that be worth the effort.

Doesnt seem that hard with the 3/4 9/12 coil-magnet-ratio
*
Steadfast
« Reply #23 on: March 15, 2012, 08:40:48 PM »
Ummm.. you guys are hurting my electicity ignorant brain.... I am an expert and even savant in many medias and art forms, but I am a dolphin at a mountian climbers convention when you guys start to talk electric tech like this... please speak to me about electricity and circuits as if I am much more mentally challenged...
See, look under my name... it says "newbie" for a reason...
thanks...
*
tecker
« Reply #24 on: Today at 04:47:42 AM »
Ty the gate to the anode with a signal diode for half wave Rectifier
***
~~~~~~~~~~~~~~~~~~~~~~~~~~~
Beginning of thread----

I decided ok, lets check that out, ie: "tie the gate to the anode" I was pretty optimistic, a diode in series is a halfwave rectifier.
As Dinges pointed out a shorted SCR cant work, he is correct. When the gate is tied to the anode, the SCR conducts in both directions irregaurdless (like its shorted).
What I did find that was interesting, was that a led in series with the connection (gate to anode) caused the SCR to act like a diode. Im pretty sure the arrangement will need a shunt resistor in parallel to protect the diode(led). Im also considering trying other diodes like a 4011 or IN914.








I just find this interesting.

[JW]

Here's a simple schematic of the connections

[JW]

I bought this book on SCR's on ebay today for $3.00 usd


I can get the 125 amp 200volt SCR for about $8.00 usd for one(new old stock) and 6 are needed (for 3phase rectifier)

Also in My research I found this 3ph rectifier for less than $18.00 usd, hard to beat that price
http://www.amazon.com/80-Amp-3-Phase-Bridge-Rectifier/dp/B004FGIVN8

[tecker]

Where did you find the 200 volt 125 amp scrs

[JW]

There not easy to find at that price, you have to hunt for them. You will find alot of stuff for about $40.00 apiece. This isnt the exact one I was thinking of but it shows the price of these can be way less that $10.00  I have found them as low as $4.00

550416 AMP 200 VOLT SCR (SILICON CONTROLLED RECTIFIER)-Igt: 30ma Vgt: 1.5V Vf (on): 1.6V-TO-48 PACKAGE $9.83

http://www.electronicplus.com

[tecker]

Ok I'll check ebay from time to time . I'm leaning toward a PnP NpN pair as a rectifier to allow low speed charging some chance . I think a closed loop with a sensing coil is a just what the doctor ordered . i just picked up some mj2955s to set up a rectifier with 2n3055 .
The idea is to be in time with the stator and conduct hard when the rotor drops in the conduction hole and release when the rotor would normally Buck .

[JW]

I did recieve the GE book on SCR's that I ordered off ebay for $3.00

Here are some pictures-









The book seems pretty good, I was thumbing thru it the other night, cant find the same circuits I was thinking of, but this one is close. :ie 2 SCR's and 2 Rectifier diodes make up the bridge.

Im really interested in casting a "rectifier coil" in the stator im working on. I have 8 magnets 2in by 2in by .5in thick.

Im going to use dual dotors with three coils in the stator. Not sure of the best prop size, but thats the starting point...

-edit-The single phase rectifier circuit wont work for my application, it will have to be a 3phase rectifier-

Ya, if your watching ebay alot,  the "new old stock" is sold for dirt cheap, have to be in the right place at the right time I guess.

[definitionofis]

Here's a simple schematic of the connections

(Attachment Link)

I joined fieldlines just so I could post a message to say I think that diode, which is tied to the SCR gate, is backwards.

I am using some old SCRs as diodes and I was hunting the internet to confirm my idea.
I tested at less than 6 volts, without a gate diode. I am just tying gate to anode). The idea works.
 
I added the gate diode so I can block negative gate voltage so as to not harm the gate. Thanks for the idea.

Coincidentally, I pressed NEXT and the next thread that came up was about LTC3703 as a buck converter for MPPT for a wind turbine.
I'll post about my LTC3703 project on that thread.

[JW]

Bump

[JW]

bump

[joestue]

I'm not sure what the point is. they have higher forward voltage drop, and unfortunately the biggest issue with using them in a wind turbine situation is the power factor or rms input amps get really excessive as the firing angle drops out.

basically because current is conserved and there is no energy storage network, the rms amps flowing into your battery are also the rms amps flowing through the alternator.

where as a buck converter with a modest input filter (you have to have one, or the switches blow up) the output amps can be far higher than the rms input amps.

you can use a big LC filter on the input of an SCR controlled rectifier, but because the wind turbines are relatively low frequency, the energy storage requirements are very high and you need a big expensive inductor.


simply using 6 mosfets as an AC bidirectional switch to short out all 3 phases together prior to the rectifier (driven from just one gate drive transformer).. can be used to make a boost converter using the inductance of the alternator. then make the alternator bigger and sized to reach full load without the help of any voltage boost just before furling..

you then rely on the boost function to get power out of the turbine in lower winds. you don't even really need to use high quality diodes for the rectifier, if the frequency is low, like say below 10 khz.. although it might be better to buy fast diodes.