Duel variac setup...

Discussion in 'The Projects Forum' started by ramblin, Jun 15, 2016.

  1. ramblin

    Thread Starter New Member

    Jun 15, 2016
    12
    0
    I am currently working on a project where I need variable power between 200 - 250vdc @ 4 - 8 amp range. I have (2) 0 - 140vac @5amp POWERSTAT variable autotransformers. Is there any reason I can not use these together, one each on each leg of my 220ac line connected to my 1000volt-50 amp FWRB powering an inductive load. Would I experience any weird feedback or need any extra components in the circuit to prevent problems ? I have been using 2 cheap Ebay SCR type (triac-diac) type controllers wired in this manner but control is less the desirable and occasionally I fry a triac. Any assistance in improving the control and reliability of this circuit would be greatly appreciated.
     
  2. GopherT

    AAC Fanatic!

    Nov 23, 2012
    5,984
    3,722

    What was the max (peak) voltage of the scr? What did you calculate as the peak-to-peak max of your 120+120 vac circuit?

    Why do you think adding two autotransformers, would be more stable than the SCR option?
     
  3. Lestraveled

    Well-Known Member

    May 19, 2014
    1,957
    1,215
    I don't see any problem in what you are thinking about doing. You are not putting them in parallel so balance is not a problem. It should work.
     
  4. ramblin

    Thread Starter New Member

    Jun 15, 2016
    12
    0



    The SCRs are stable if amps are held below 4-5 amps. Voltage will go as 205 max immediately (which is fine for my application) but amps are hard to control and once you pass 4-5 amps it will latch and go to max (7 -10 depending on my config. Even prolonged running at 4-5 amps the triac-heat sink will get extremely hot and eventually fail. Just not good control. Just not designed for inductive loads I guess. That is my question; Will 2 autotransformers be more stable and reliable. I already have them and don`t really want to burn them up. The 5000watt SCR controllers with 40amp- 600volt triacs are available on Ebay for about 7 bucks. The variacs, not so cheap.
     
  5. GopherT

    AAC Fanatic!

    Nov 23, 2012
    5,984
    3,722
    I'd try the $7 triacs with big heat sinks. Big, big heat sinks.
     
  6. ramblin

    Thread Starter New Member

    Jun 15, 2016
    12
    0
    I added extra heat sinks and a fan. Still get poor control and I melt the fuse holders off of them and burn triacs. Do you not think that variacs are a good idea ?
     
  7. GopherT

    AAC Fanatic!

    Nov 23, 2012
    5,984
    3,722
    My logic is as follows:
    Background data
    - some part of your system is not well controlled and some parts are getting stressed/damaged.
    - you don't want to damage two expensive variacs
    - but triacs are available for $7

    It seems to me that a triac is the best option to try (doesn't damage triacs, may not get damaged if new part meets demands of circuit.

    - work with it until you discover it doesn't work well, the. Try another option but the
     
  8. ramblin

    Thread Starter New Member

    Jun 15, 2016
    12
    0
    My logic is as follows:
    My inductive load varies widely thru out my testing process. I am but a rookie but I have asked questions other places and been informed that SCRs are NOT designed for an inductive load and been advised to use a variac for that reason. That is why I am asking advice on how to best protect them with additional components. A good 220volt variac can cost hundreds and I already have (2) 120v 5 ampers. And keep in mind I still have poor control with the SCRs even though they are cheap.
     
    Last edited by a moderator: Jun 16, 2016
  9. ramblin

    Thread Starter New Member

    Jun 15, 2016
    12
    0
    2006-05-22 13.26.01.jpg This is the cheap SCR I been using. I would still like to know if I need to add anything else to the circuit to protect my variacs.
     
  10. ramblin

    Thread Starter New Member

    Jun 15, 2016
    12
    0
    Last chance for any advice. Got my panel meters today . Going to build my power supply and put it in a nice case. Don`t want to burn up anything. ??
     
  11. ramblin

    Thread Starter New Member

    Jun 15, 2016
    12
    0
    OK, I have completed my duel variac project as seen here. Each one works fine on its own, complete with amp and voltmeters, switch and fuse. Truth is I haven`t got a clue how to hook up a FWBR. Obviously I can`t just hook each one to a single bridge with out a ground. Do I use 2 bridges ? I can`t just add one on to the other can I ? At least one person here thought I could make it work. All I want to do is get 0 to 200 vdc. Can someone tell me how to hook this up ? 2006-06-05 18.07.25.jpg 2006-06-05 18.09.42.jpg
     
  12. KeepItSimpleStupid

    Well-Known Member

    Mar 4, 2014
    1,136
    200
    So, your like in a US based system with a center-tapped 120-0-120 service? My answer will assume that,

    Remember with ANY VARIAC, the wiper MUST be fused. The input can be as well. The maximum current output has to be limited, not the input current. You should also disconnect both legs simultaneosly (Again assume US type of system)

    if you need isolated power, then at least one isolation transformer would be required. You can then, and only then, connect any side of the output to ground if required.

    You'll get about 165 ish DC from full wave rectified 120 VAC and one Variac. Take a look here: http://www.hammondmfg.com/pdf/5c007.pdf There are issues with the currents required for the transformer vs the DC currents. Pay attention to the DC and AC currents in that figure.

    You can arrange the variacs to provide CC increasing voltages, essentially out of phase.

    1000 uf/AMP is the minimum capacitance I would use.

    Two Variacs and a 240:240 isolation transformer will give you about 0 to 320 VDC. I DC = 0.62 of I AC, so there is an issue there, The xformer output has to be greater than 1/0.62 * 8 Amps. That isolation transformer would be connected between the two wipers of the Variacs,

    You can use another transformer to "subtract" an AC voltage if needed if you truely want to limit the voltage to around 250 VDC,
    That can be figured out. It's also in the PDF. You can also subtract DC, possibly.

    If the desired output is DC, Why didn't you use a DC meter? (Anyway: I accidentally read that you needed 350 VDC (My Oops).

    I think you assumed that full-wave AC in rectified and filtered is DC out and it isn't. It's the peak value. Inductive loading adds yet another complication. The system should be sized for max VA or Volts*Amps (Volts and Amps Separately). That would be more of instantaneous Amps.

    I think you got led astray, Hopefully I didn't make any major mistakes. In the right place you can add or subtract AC voltages. We MAY have to discuss phasing.
     
  13. ramblin

    Thread Starter New Member

    Jun 15, 2016
    12
    0
    Let me attempt to clear up a few things. Yes I am US based with approx. 120vac each phase. I am working with parts I already have (2) 120 variacs. My goal is controlable 0 to 200 plus DC and upper limits is not an issue. Right now each variac with meters est. are wired seperatly and no attempt to connect them together yet. I realize that connecting the 2 phases together could blow everything and that is why I am here asking advice. Its not safe to assume I know any more than that because I don`t. The FWBR and smoothing cap are mounted but not connected. Are you saying I need an isolation transformer between to make this work ? I do have (1) 120vac isolation transformer I could use if that would work. I also have (2) other 220vac multitap transformers but they are not isolated. They maybe could be made to be if I could use them. This has been my main point about this thread is how to tie these 2 variacs together and make it work.
     
  14. KeepItSimpleStupid

    Well-Known Member

    Mar 4, 2014
    1,136
    200
    They are connected together, by the nature of the neutral.

    If you connect the bridge rectifier to the two wipers and the DC output of the bridge to a capacitor.

    You WILL get a variable DC voltage to at least 1.4*240-2(0.6) VDC.

    The problem is, you CANNOT connect either end to ground.

    If you want to fix that, you need an isolation transformer. 240:240 between the two wipers of the variacs and the DC output of the bridge rectifier.
     
    Last edited: Jul 1, 2016
  15. ramblin

    Thread Starter New Member

    Jun 15, 2016
    12
    0
    I just can`t see how the isolation trans. is wired in. Maybe you could draw something up ? Say if I`m powering a simple coil from the bridge output. Ground goes to one side of the bridge input. So the isolation transformer must combine the two outputs from the variacs to one wire to be connected to the bridge input ? Is that correct ? How ?
     
  16. KeepItSimpleStupid

    Well-Known Member

    Mar 4, 2014
    1,136
    200
    Totally forget about ground or neutral except make sure your external circuit doesn't use it.

    Is the bridge right now, being driven by the two wipers? I think it is, but it's difficult to follow the wiring.
    Therefore you should have variable DC across the capacitor.

    If you have any kind of resistor that satisfies the power rating for say 400 V (P>(400^2)/R), do it. It will bleed the cap, otherwise it will appear "stuck" or the voltage will only go up and it will go down very slowly.

    The idea being, is that when you turn it off, you will need 5 time constants (5*R*C) to be 99% discharged.
    So, pick a time with your value of C in Farads. e.g. 100uf = 100e6 Farads. Farads*Ohms = time. Time will be in seconds.

    So, pick a reasonable time, like 30 seconds maybe.

    Right now, a DC voltmeter across the cap IF you have it wired as wipers only to bridge AC inputs, the DC voltage might not go down very fast when you turn the variac down.

    Two 120 light bulbs in SERIES will work. Something like 7W INCADESENT night lights in SERIES.

    If you don't have an external grounded system, the setup will work. The Variacs could subtract if they are wired wrong.

    So, you really need a load, but you can take a look at it as is. Raising the AC voltage will work, but lowering it may not immediately lower the DC voltage without a load of some sort.

    Really really busy today especially.
     
    Last edited: Jul 2, 2016
  17. KeepItSimpleStupid

    Well-Known Member

    Mar 4, 2014
    1,136
    200
    Isolation (quick and dirty)

    You realize that the house is being fed from say a 10 kV primary and a 120-0-120 secondary and the 0 point is grounded?

    You get 240 VAC when you connect to both ends. Ground is there for safety and you can't connect L1 or L2 to ground.

    Something that MIGHT help you understand what's going on in a typical L1, L2, N and G type circuit is:
    If L1 and L2 have identical loads, the current in the neutral is zero. When unbalanced, the N carries the difference current and it can change sign. G and N are connected together at one single point in your house.
    G either provides a reference or conducts a fault current. When a fault occurs with outlets wired in parallel, it can upset the ground reference when a fault occurs.

    You have 120 V Variacs, so you can only connect them one way to a 120-0-120 service. One to L1&N and the other to L2 & N. So, N is common,

    Your wiper to wiper AC output will either add or subtract when the Variac is turned Clockwise based on how the Variac is wired It will probably add which is good.

    So, your just creating your own custom transformer as you move the dials: e.g. a 30-0-24 or 100-0-36 or even a 100-0-0 transformer. 0 is ground/Neutral. I think you can understand that. The reason why the voltages add is that they are out of phase.

    A water heater uses a 240 resistance element and needs no other connection to ground,. If your load is like this your good.

    the new dryer wiring and even stove wiring will now likely have L1, L2, G and N. A small amount of L1 to N or L2 to N (120 VAC) is used to run the controls and it keeps the safety stuff (protective ground) in place.

    You could, but I would not do it is to isolate the 240 coming in. You could isolate each 120 V Variac which is better.

    The best option is to isolate the output of the wiper to wiper system.

    Using a 100-0-50; labeled a,b and c. The isolation xformer primary would connect to a and c and your isolated output would be the secondary of the isolation transformer. So, the primary sees 150 (100+50) Vac and the secondary sees 150 VAC and you removed the connection to ground. The AMOUNT of isolation is actually determined by the ISOLATION value of the transformer. e.g. 3 kV.

    A 240:240 isolation xformer will work because the ratio is 1:1 AND the max input is 240 V. You could very well use a 460:460 transformer there as well, but a 460 VAC xformer costs more.

    You would need four 120:120 transformers to also accomplish what you wanted to do. (discussion postponed).

    You could also use say two 240:24 volt transformers as well. You can arrange them as 240:24 feeds 24:240.

    Phasing is important as well as the current rating of the primary/secondary.

    So, power is conserved. Pin = Pout, but the Max V and Max I have to be adhered too.
    Unfortunately, they are not PERFECT transformers, so there is 'regulation" and some "losses". Small, but they are there,

    When you use the named secondary as the primary there is a slight efficiency penalty because the new "primary" is further from the transformer core.

    I hope this explanation is better?
     
  18. ramblin

    Thread Starter New Member

    Jun 15, 2016
    12
    0
    KISS, I followed everything you said in #16 and all works exactly as you said it would. To be perfectly honest I never knew the circuit would work like that without a ground. And my purpose in this thread was to learn without burning up my stuff. I had only tested it without a load. And I appriciate the lesson in isolation. To most on this forum this is kindergarden stuff but if no one takes the time to explain it can be a little hard to wrap your head around. You can only learn so much studying and reading the forums. Thanks to everyone for all your help. Been having a blast blowing ot 100 watt bulbs all afternoon. It is the 4th of July !
     
  19. KeepItSimpleStupid

    Well-Known Member

    Mar 4, 2014
    1,136
    200
    Great! When you start mixing things like (electrical and electronics) it's no longer kindergarten.

    Most are dealing with regulated DC. It's somewhat unique. All basically because you had 120 VAC Variacs.

    When the "supervisors" moved us to a building made for us, they really messed up. Some labs had outlets every feet (good), some didn't. One guy secs some isolated ground stuff. I would have made the 120 outlets higher off the ground. A lot of equipment needed to have 208 4 wire and they totally missed that, They also intermixed hallway and la outlets, so a polisher could ruin a lab experiment.

    One of the stupider things they did was to cool equipment that is designed for groundwater cooling off a 100 degree cooling loop. Stupider yet they tried to use a booster pump to pump it through a 1/4 line. Fortunately, they plumbed drains into the labs.

    They also put two heat pumps (water to air) on different thermostats in the SAME room.

    The builders had a unique way of building an isolated concrete pad. Pour all of it, then break out the isolated portion and pour again.

    Later, I discovered that a lot of the outlets were defective from the manufacturer after some equipment started dying. A very strange problem: When you plugged two pieces of equipment in a duplex outlet, one MIGHT loose a ground. Identifying the ad outlet was difficult. Electricians clueless. Potentially, there were 470 outlets where some were defective and some could do some serious damage.

    On one of the re-models, they plastered valve handle into the wall so it was not movable. One problem really stumped the HVAC guy. He diagnosed the issue as a bad heat pump. I saved his butt with about 10 minutes of my time. I asked him to do a few measurements and told him to replace the t-stat wire. The re- modeler's skinned the t-stat wire on the metal studs without grommets. He was apprehensive, I was totally sure.
     
Loading...