A Hendey 9 T&G "wreck" - Page 2
Close
Login to Your Account
Page 2 of 9 FirstFirst 1234 ... LastLast
Results 21 to 40 of 179
  1. #21
    Join Date
    Dec 2002
    Location
    Monterey Bay, California
    Posts
    10,260
    Post Thanks / Like
    Likes (Given)
    28
    Likes (Received)
    217

    Default

    "So I guess this is some sort of current regulator. Yes? No? Anybody?"

    Without having any "active"/"non-linear" devices, the simple armature and field regulators which I described earlier would have no load regulation.

    I believe the saturable reactors are an attempt to add I-R compensation ("I-R comp") to the drive.

    The armature regulator certainly needs I-R comp, the field regulator less so.

    In the Monarch WiaD and Modular, the I-R comp circuit applies only to the armature regulator.

    In the Monarch M-G, the generators are compound wound, and this, then, provides some measure of I-R comp. The Monarch M-G drive has a compound wound exciter (and which supplies the field through a "lossy" rheostat) and a compound wound generator (and which supplies the armature). The generator's field is also supplied through a "lossy" rheostat, but the main generator is connected directly to the spindle motor through the reversing contactors.

    Monarch's patent on its WiaD would have lasted until long after the last T&G Maker's lathe was made.

  2. #22
    Join Date
    Oct 2003
    Country
    CANADA
    State/Province
    Ontario
    Posts
    1,178
    Post Thanks / Like
    Likes (Given)
    71
    Likes (Received)
    121

    Default Interesting find

    While taking some detailed pictures today for Cal, I came across a painted over data plate in the control cabinet, that I had not paid much attention to before.



    It appears that the drive control system for my lathe may have been built by Louis Allis company.

    Does anyone know if they still exist, or if there is any successor company that might have information?

    Cheers
    Pete

  3. #23
    Join Date
    Apr 2002
    Location
    Central Mother Lode, California
    Posts
    807
    Post Thanks / Like
    Likes (Given)
    973
    Likes (Received)
    100

    Default

    Quote Originally Posted by shapeaholic View Post
    While taking some detailed pictures today for Cal, I came across a painted over data plate in the control cabinet, that I had not paid much attention to before.



    It appears that the drive control system for my lathe may have been built by Louis Allis company.

    Does anyone know if they still exist, or if there is any successor company that might have information?

    Cheers
    Pete
    My 1977 10EE had as its original motor a 5 h.p. Louis Allis D.C. motor. If you Google "Louis Allis" you will find they moved from Milwaukee to to the south east U.S., and apparently are still in business.

    The following tells about the company falling on hard times in the late 1990's (before their move to Alabama).

    Louis Allis: What happened? | The Business Journal

    David
    Last edited by old_dave; 01-08-2011 at 05:47 PM. Reason: minor corrections

  4. #24
    Join Date
    Dec 2002
    Location
    Monterey Bay, California
    Posts
    10,260
    Post Thanks / Like
    Likes (Given)
    28
    Likes (Received)
    217

    Default

    Allis-Chalmers went under.

    At one point in time, it had built the largest turbine-generator then in use. 1,100 MW ... 1,100,000 kW ... the infamous "Big Allis".

    Leeson bought what remained of A-C's small motor business.

  5. #25
    Join Date
    Nov 2005
    Country
    UNITED STATES
    State/Province
    New Hampshire
    Posts
    449
    Post Thanks / Like
    Likes (Given)
    692
    Likes (Received)
    292

    Default On photographing metal

    Taking photo's of shiny metal is difficult at best. The reflections will drive the best photographers nuts & render the best equipment useless, unless you know lighting. That is the key.

    Diffused light is part of the equation. What needs to be done in addition to that is block some of the light hitting the machine ways, while allowing the remaining light to light the surrounding areas using thin strips of black paper taped over a rod or some other method. The angle of the reflections has to be out of the path of lens. Only then will you get good results. It is not hard, but takes some experimentation and thought. If you want to learn more and like buying books, this one will explain it better than I.
    Amazon.com: Light: Science and Magic: An Introduction to Photographic Lighting (9780240808192): Fil Hunter, Steven Biver, Paul Fuqua: Books

    I am a hobby photographer so I don't mind reading books on such esoteric subjects as lighting.

    But for the most part, I find the images posted by you folks to be good enough to convey the idea. I will work on it some & see if I can get a pic of my set up. You should be able to get good results with a point & shoot and some basic lights you already have with some care.

    Warren

  6. #26
    Join Date
    Dec 2002
    Location
    Monterey Bay, California
    Posts
    10,260
    Post Thanks / Like
    Likes (Given)
    28
    Likes (Received)
    217

    Default

    It would indeed be interesting to have a copy of the drive calibration procedure for the B-C non-electronic drive.

    There are two adjustable resistors which can affect the behavior of the drive: R1 and R3. The remaining resistors are fixed.

    The drive appears to have two approaches to load compensation: 1T1 in parallel with 1T2 but shunted by R1 affects the potential across 1T2, most probably bucking the armature voltage when under low loading; whereas 1SR1 in parallel with 1SR2 but not shunted by anything affects the potential across 1SR3, most probably boosting the armature voltage when under high loading.

    An overload in the armature circuit will pop overload M, thereby shutting down the drive.

    A failure in the field circuit will be detected by normally open (but closed after power-on) FFR, thereby shutting down the drive.

    If neither F nor R are selected, M-4 will be closed and the field will be set to maximum, thereby applying maximum braking power to the spindle.

    An innovative, but simple drive.

  7. #27
    Join Date
    Oct 2003
    Country
    CANADA
    State/Province
    Ontario
    Posts
    1,178
    Post Thanks / Like
    Likes (Given)
    71
    Likes (Received)
    121

    Default The scraping plan

    Well. I think that I have my scraping plan figured out.
    The first step was to make a gage to keep the geometry of the tailstock way correct. This was a bit of a chore as the shape of the ways made for some interesting workholding to be able to get the scraper into a usable position.




    Next job will be to scrape the tailstock ways straight down about .003" which will remove the damage about half way down the bed.
    After I get that done, I will work on carriage ways.
    My initial plan is to use some sort of sled and plane the way to remove the damaged areas, then align and finish the ways.
    I have given some consideration to grinding the carriage ways after I plane them as I am considering moglice on the carriage. I have used it a little before and found it pretty easy to do once the alignment was established. I am a little concerned how it would work on a scraped surface rather than a ground surface.
    Can anyone comment on that aspect?

    While I was working on the gage I did try to scrape the bed to see just how hard it was. I can assure anyone who questions it, that Hendey made the bed "HARD". I used both an Anderson carbide blade and a Sandvik insert (which I find quite superior) and both required a lot of down force to cut. I hope that subsequent cycles will be a little easier as the surface wont be quite as slick as the first pass.

    I'll see how the tailstock way goes. If it is too hard to work with I may park the project until the bank account matches the estimate from the grinder

    More to follow ;-))
    Pete

  8. #28
    Join Date
    Nov 2007
    Location
    Portland OR
    Posts
    239
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    0

    Default

    I am still curious why you made a gage instead of turning the base of the tailstock into a template. Is it because you may use Moglice on the tailstock slides?

    What surface did you use for leveling?

  9. #29
    Join Date
    Oct 2003
    Country
    CANADA
    State/Province
    Ontario
    Posts
    1,178
    Post Thanks / Like
    Likes (Given)
    71
    Likes (Received)
    121

    Default

    Cecil,
    I thought about using the tailstock base, but decided on the gage because:
    1) the tailstock base is worn,
    2) I was trying to preserve the geometry as best I could. Please note that the gage has a "lug" that fits down between the flats below the tailstock "V". I am trying to ensure that I get "straight down".
    I may be misguided but better more caution than less.
    besides I need all the scraping practice I can get. Right?

    For levelling I used the least worn portions of the tailstock slideway. I placed my 36" parallel straightedge on there and put the level on that. To level crosswise, I used the tops of the carriage V's at the extreme ends as they look to be the least worn.
    I also checked this against the portion of the bed under the headstock.

    Pete

  10. #30
    Join Date
    Sep 2002
    Country
    UNITED STATES
    State/Province
    Arizona
    Posts
    3,631
    Post Thanks / Like
    Likes (Given)
    828
    Likes (Received)
    484

    Default

    Quote Originally Posted by peterh5322 View Post
    ...
    There are two adjustable resistors which can affect the behavior of the drive: R1 and R3. The remaining resistors are fixed.

    The drive appears to have two approaches to load compensation: 1T1 in parallel with 1T2 but shunted by R1 affects the potential across 1T2, most probably bucking the armature voltage when under low loading; whereas 1SR1 in parallel with 1SR2 but not shunted by anything affects the potential across 1SR3, most probably boosting the armature voltage when under high loading.

    An innovative, but simple drive.
    Peter,

    I don’t think that’s right. The “primary” windings of 1SR (1SR1 and 1SR2) are connected with opposite polarity, so 1SR can’t act as a transformer and produce significant voltage on the “secondary” windings, 1SR3. Plus it doesn’t make sense to couple AC voltage in after the rectifier. Here’s a section of the diagram with the polarity of 1SR and 1T windings marked with blue and red dots, respectively:

    (You can verify the polarities by reference to the left half of the diagram which I posted earlier.)

    This system does make sense when you treat 1SR as a saturable reactor (SR): 1SR3 is the control coil for the reactor. When the DC current through 1SR3 is small the inductance of coils 1SR1 and 1SR2 is large; the inductance is probably arranged so that the voltage on the primary windings of 1T (windings 1T1 and 1T2) is near zero when the 1SR3 control current is low. When that’s the case the 1T secondary voltage (on winding 1T3) will be small as well. 1T3 is wired so that it reduces the voltage provided to the rectifier by the variable auto-transformer. Note that the DC armature current runs through 1SR3. As the current through 1SR3 increases it pushes the core of the SR further into saturation, causing the inductance of 1SR1 & 1SR2 to decrease, which increases the voltage on 1T’s primary. That in turn increases the voltage on 1T3 and reduces the voltage applied to the rectifier. So the system provides negative feedback, reducing the armature voltage when the current tries to increase. Increasing the voltage on the variable auto-transformer changes the set point for the armature current. Resistor R1 appears to allow some sort of tuning of the feedback loop. The SR is probably designed so that 1T’s primary gets nearly the full bus voltage when the armature current is maximum.

    This military tech. manual does a nice job of explaining SR’s and magnetic amplifiers. It has an example of using a magnetic amplifier to control a single phase AC motor (see pages 43-45):
    Servo Systems and Data Transmission
    by U.S. Dept. of the Army, U.S. Dept. of the Air Force
    TM 11-674 / TO 16-1-277
    This patent covers a 3-phase magnetic amplifier used to provide power and control for a DC motor without using rectifiers:
    The more I study the single-phase Hendy design, the more I am impressed by it’s sophistication. It would be interesting to hear to well it performs. Pete’s machine has a 3-phase version of the magnetic amplifier drive which is much more sophisticated than the single-phase version. It includes 6 large saturable reactors, an interesting little magnetic amplifier module, and a bunch of selenium rectifiers.

    Cal

  11. #31
    Join Date
    Nov 2004
    Location
    Webster Groves, MO
    Posts
    7,272
    Post Thanks / Like
    Likes (Given)
    1826
    Likes (Received)
    3318

    Default

    Cal,

    Saturable reactors frequently have a three legged core with the two outer legs carrying the current being controlled. A common configuration is to send the current in one coil, through the device being controlled, and out through the second coil, although there are countless variations. With no DC on the center coil, the two outer legs act like chokes sharing a common center leg. Applying DC to the center coil effectively removes the iron from the outer coils. The nice thing about this configuration is that the two AC fluxes cancel and only a little noise and harmonics appear on the center coil. That makes life a lot easier for the control current source. This is especially helpful when you are trying for a large current gain. You need as many ampere turns on the control winding as the other two. That means the control winding will have many more turns to get the gain. The power gain comes from having only the DC resistance of the control winding with inductance not a factor. Without this cancellation, it would act as a transformer, generating an AC voltage in the control winding. With only a 10:1 current gain on a 120 VAC reactor, you would have to deal with 1200 VAC coming from the control winding. On something like the 240 V, 3 phase saturable reactor controller we made for my 14 1/2" South Bend, the control transistors would have a rough time. Instead, we just have a little high impedance crud that a load resistor swamps easily. BTW, this is pure reactor control, no autotransformers or whatever. I have three ten turn pots for speed controls with pushbuttons like an auto radio to select them. Roughing, finish cut, polish, just push the button.

    It looks to me that they have paralleled the two outer windings on 1SR for the same cancellation effect and 1SR3 is the control winding. Probably 1T is a simple transformer that adds voltage to the output of the autotransformer for IR comp. I don't know why it has two primary windings if that is in fact the function.

    The biggest problem with the standard three leg design is leakage reactance, magnetic lines jumping between legs instead of following the core, which effectively decouples some of the flux from the coils. Stacked toroid cores pretty well eliminates the problem, but they are difficult to wind. I have a more developed design using two rectangular cut cores cut on the short leg instead of the long one. I wound individual power windings on the legs, stacked two together and wound control windings over them. Then I strapped the cores together, giving me four power windings with two control windings over them. I use them on single phase to replace the Reliance MG on my 10EE. I have it set up with a 5 hp DC motor used as a generator for a load on the spindle and can easily run it over the speed range with any load I choose. The main problem has been getting IR comp that works at all speeds and loads. I haven't done anything with it lately because economic survival takes precedence.

    Bill

    Edit. The thing I like about reactors is that they are nearly indestructable. By limiting the control current, you limit the output current as well. I can hold in the spindle lock on my 10EE and throw the power on at the full speed setting. The current just goes up to 100% and sits there for as long as you like. It does overheat one set of armature windings, so it isn't an indefinite time, but it also limits the starting surge to 100% current.

  12. #32
    Join Date
    Dec 2002
    Location
    Monterey Bay, California
    Posts
    10,260
    Post Thanks / Like
    Likes (Given)
    28
    Likes (Received)
    217

    Default

    Thanks for setting me straight, Cal.

    Definitely an interesting drive.

    Peter

  13. #33
    Join Date
    Jul 2010
    Location
    Rochester, NY USA
    Posts
    2,172
    Post Thanks / Like
    Likes (Given)
    10
    Likes (Received)
    507

    Default

    Pete,
    Nice photos of the electrical cabinet contents would be useful as the discussions about how this control works continue. And very helpful in finding one similar to yours.

    Bill

  14. #34
    Join Date
    Oct 2003
    Country
    CANADA
    State/Province
    Ontario
    Posts
    1,178
    Post Thanks / Like
    Likes (Given)
    71
    Likes (Received)
    121

    Default

    Bill,
    Here is a link to the album with a bunch of detail pictures of the drive components.

    hendey drive pictures pictures by shapeaholic - Photobucket

    more to follow

    Pete

  15. #35
    Join Date
    Nov 2004
    Location
    Webster Groves, MO
    Posts
    7,272
    Post Thanks / Like
    Likes (Given)
    1826
    Likes (Received)
    3318

    Default

    That is a lot of copper and iron. One thing about it, you can always scrap it for some quick beer money.

    1T and 1SR both seem to have dual cores and are actually two separate units in each. Since the current passing through 1SR3 is probably higher than the currents in 1SR1 & 2, the control winding probably has fewer turns than the power windings so the voltage induced in it would not be troublesome. I think if you trace the circuit out, you will find that windings 1SR1 & 2 are each on a core and 1SR3 is windings on both cores in series so the induced voltages buck. One of the problems with a simple saturable reactor is that the DC bias makes it saturate quicker in one direction, making the output a little unsymmetrical. I would bet that these are set up with the dis-symmetries reversed to give a better waveform. Furthermore, having the control windings on top of the power windings instead of on another leg minimizes the leakage reactance problem and still makes a simple, easy to manufacture unit. Good practical engineering.

    That leaves the mystery of why 1T is two units. I can't read the nameplates for certain, but they appear to be Hevi-Duty brand. They would have had any imaginable core available, so there would be no incentive to use them to reduce inventory. The best explanation I can suggest is that 1T and 1SR are the same, just connected differently. They might have gotten a better price for the larger quantity. That is a bit of a stretch, but I can't think of another reason.

    I would really like to see the 3 phase system. I have made several 3 phase saturable reactors with good results.

    BTW, the terms saturable reactor and magnetic amplifier are used almost interchangeably, but should not be. To be a real magnetic amplifier, there has to be a feedback loop that makes the unit bootstrap itself, reducing the control winding energy required or some similar gain producing circuit. Mag amp sounds cooler, though, so it gets misused.

    Bill

  16. #36
    Join Date
    Sep 2002
    Country
    UNITED STATES
    State/Province
    Arizona
    Posts
    3,631
    Post Thanks / Like
    Likes (Given)
    828
    Likes (Received)
    484

    Default Hendy 1-ph vs. 3-ph "magnetic amplifer" drives

    Quote Originally Posted by 9100 View Post
    ... 1T and 1SR both seem to have dual cores and are actually two separate units in each. Since the current passing through 1SR3 is probably higher than the currents in 1SR1 & 2, the control winding probably has fewer turns than the power windings so the voltage induced in it would not be troublesome.
    The photos of Pete’s (3-phase) machine and the single-phase diagram don’t go together and the two designs don’t seem to have a lot in common (apart from the component designations). If you look at the left half of the single-phase diagram:
    1T and 1SR are each drawn as 3 coils on common cores (one core for 1T and another for 1SR). The “secondary” coils are drawn with fewer turns than the “primary” coils, as you surmise. The “Electrical Stock List” in the upper left-hand corner of the diagram lists the “Transformer Coils” used in 1T and 1SR. A total of 4 of the same part number coils are used for the “primary windings”, 1T1, 1T2, 1SR1 and 1SR2; similarly, 2 identical coils are used for the secondary of 1T (1T3) and control coil for the saturable reactor (SR) 1SR3. There’s no way of telling from the diagram if the cores for 1T and 1SR are the same. From what you say below, it sounds like it might make sense to use the same core design for both 1T and 1SR?

    Given that the “primary” and “secondary” coils of 1T and 1SR are the same, would you expect the voltage across 1T’s primary to be about 50% of the bus voltage (110VAC bus) when the control current to 1SR3 is near zero and near 100% when the control current is at maximum? (I guess the armature current would never be zero, would it—what sort of minimum current would you expect given a 2HP motor idling a minimum speed?)

    On the other hand, the SRs in Pete’s 3-phase machine (a total of 6) are each wound on a separate core and mounted in pairs. 1SR and 2SR are mounted together. Pete tells me that there are 6 wires going to each SR coil; 2 are of a heavier gauge than the other 4; apparently there are three windings on each core. The cores for the SR’s are rectangular, with a hollow center. The coils are wound on one side of the core and fill the center of the core. Pairs of SRs are mounted with the windings to the outside with no contact between the cores, apart from the mounting bars.

    The construction of Pete’s SRs appears to be similar to the current transformer on the main panel, 1CT. It has the same core geometry as the SRs. It appears that 1CT’s cores are made up of U-shaped laminations that are slid through the center of the coil, with adjacent laminations coming in from opposite ends to form the rectangular core. Pete says he can’t tell if the SR cores are built the same way, but I would think it’s likely; what do you think?

    Here’s a couple of photos that show 1CT:
    TB3(?) and 1CT (The photo is rotated 90 degrees; the TB should be vertical. 1CT has a blue tag.)
    top of 1CT (lower left in frame)
    Pete has uploaded higher resolution (1600x1200) photos of his panel to this sub-album:
    (I’m posting links to the photos, rather than placing them inline, due to their size.)

    Quote Originally Posted by 9100 View Post
    I think if you trace the circuit out, you will find that windings 1SR1 & 2 are each on a core and 1SR3 is windings on both cores in series so the induced voltages buck.
    Back to the single-phase diagram: As I mentioned above 1SR and 1T are drawn as three windings on a single, vertical cores. Maybe the single-phase design’s 1SR is like one of Pete’s six SRs? If the core of the single-phase SR (1SR) has a the common figure-8 geometry, it’s not reflected in the diagrams.

    Quote Originally Posted by 9100 View Post
    One of the problems with a simple saturable reactor is that the DC bias makes it saturate quicker in one direction, making the output a little unsymmetrical. I would bet that these are set up with the dis-symmetries reversed to give a better waveform. Furthermore, having the control windings on top of the power windings instead of on another leg minimizes the leakage reactance problem and still makes a simple, easy to manufacture unit. Good practical engineering.
    Maybe that would explain why there are 3 pairs of big SRs Pete’s machine instead of 3 individual SRs. If the SR’s in each pair were wired back to back, would the asymmetries cancel?

    The small “magnetic amplifier” 4MA is shown in this photo:
    It appears to have 2 SRs of similar construction to the big SRs, with cores and windings similarly configured. 4MA differs from the big SRs in that there are connections between the coils; there are no direct connections between the big SRs, any interconnections appears to be handled on terminal strip 1TB.

    It’s interesting to note that the small “magnetic amplifier” is marked 4MA. All of the component designations on Pete’s 3-phase panel and on the single-phase diagram seem to in the form of a digit followed by one or two letters that indicate the type of component; i.e. 1R, 2R, 3R, etc. are resistors; 1T, 2T are transformers, and so forth. The digits appear to be assigned serially, starting at 1; when there is only one of a particular component, like the field loss relay, they don’t use the digit; thus, the field loss relay is simple designated “FL” (not 1FL).

    So, if we have a 4MA, there should also be 1MA, 2MA and 3MA somewhere, but those labels don’t exist. I’m guessing that 1MA consists of 1SR and 2SR; 2MA is 3SR and 4SR; and so forth. I assume that the SR suffix means “saturable reactor” and MA means “magnetic amplifier”. Pete tells me that a cursory examination suggests that 4MA is part of the field circuit and 1SR through 6SR part of the armature circuit. 4MA seems to connect to the smaller rectifiers, located above the big SRs. The big SRs connect to the bank of large rectifiers, via three bundles of wire that pass through the bulkhead between the SRs and the rectifiers.

    Quote Originally Posted by 9100 View Post
    That leaves the mystery of why 1T is two units. I can't read the nameplates for certain, but they appear to be Hevi-Duty brand.
    1T is the buck transformer on the single-phase diagram. The black transformers with the red tags on Pete’s 3-phase panel are, indeed, Hevi-Duty brand. There are actually three of them, mounted side by side. Each is a single phase, 1KVA transformer with 220/440V primaries and 110V secondary. The primaries appear to have been connected as 2:1 autotransformers and used to convert 440 to 220. The secondary of one of the transformers was used to provide 110VAC to the work light; the secondary’s of the other two are not used. Apparently the machine was 440 when Pete got it, with the transformers connected directly to the incoming power cable, ahead of 3-phase disconnect switch on the main panel. He has disconnected the transformers, connecting directly to the switch

    Quote Originally Posted by 9100 View Post
    They would have had any imaginable core available, so there would be no incentive to use them to reduce inventory. The best explanation I can suggest is that 1T and 1SR are the same, just connected differently. They might have gotten a better price for the larger quantity. That is a bit of a stretch, but I can't think of another reason.
    I believe that you’re right. As I said earlier, they use the same part number coils, so the cores are probably the same as well.

    Quote Originally Posted by 9100 View Post
    I would really like to see the 3 phase system. I have made several 3 phase saturable reactors with good results.
    Pete’s hoping he can find a wiring diagram for his panel. In the mean time, it’s very interesting to see what can be learned just by looking at it. At some point, Pete may have to trace out some or all of the wiring to produce a diagram. B-C seems to have been pretty consistent in how they numbered the wires, so it might not be too bad a job to figure out all of the interconnections.

    Quote Originally Posted by 9100 View Post
    BTW, the terms saturable reactor and magnetic amplifier are used almost interchangeably, but should not be. To be a real magnetic amplifier, there has to be a feedback loop that makes the unit bootstrap itself, reducing the control winding energy required or some similar gain producing circuit. Mag amp sounds cooler, though, so it gets misused.
    I would like to hear more about that. Could you comment on the “magnetic amplifier” shown in this military tech manual (page 45)? They discuss saturable reactors first (page 43), the move on to the “magnetic amplifier”:
    Cal

  17. #37
    Join Date
    Dec 2002
    Location
    Monterey Bay, California
    Posts
    10,260
    Post Thanks / Like
    Likes (Given)
    28
    Likes (Received)
    217

    Default

    "I would like to hear more about that. Could you comment on the 'magnetic amplifier' shown in this military tech manual (page 45)? They discuss saturable reactors first (page 43), the move on to the 'magnetic amplifier'"

    So-called "magnetic amplifiers" have been used in automatic control systems (presently called feedback control systems) since at least WW-II.

    Before the complete literature on feedback control was available, such less understood, a variety of essentially passive (that is, non-amplifying, and non-electronic) means were utilized.

    However, once Black's means and methods were widely known, and were free of AT&T's/Bell Labs' patent protection, electronically-amplified feedback-controlled systems became popular, if not absolutely essential.

    A WiaD is perhaps an example of an electronically-amplified feedback-controlled system, as applied to machine tools.

    Perhaps, it is also one of the most significant examples, too.

  18. #38
    Join Date
    Sep 2002
    Country
    UNITED STATES
    State/Province
    Arizona
    Posts
    3,631
    Post Thanks / Like
    Likes (Given)
    828
    Likes (Received)
    484

    Default

    An early magnetic amplifier was patented in 1912 by Ernst Alexanderson, radio and television pioneer:

    In 1920 Alexanderson used one as part of one of the first radios capable of transmitting voice. Alexanderson has over 340 patents to his name.

    Cal

  19. #39
    Join Date
    Nov 2004
    Location
    Webster Groves, MO
    Posts
    7,272
    Post Thanks / Like
    Likes (Given)
    1826
    Likes (Received)
    3318

    Default

    [/INDENT]1T and 1SR are each drawn as 3 coils on common cores (one core for 1T and another for 1SR). The “secondary” coils are drawn with fewer turns than the “primary” coils, as you surmise. The “Electrical Stock List” in the upper left-hand corner of the diagram lists the “Transformer Coils” used in 1T and 1SR. A total of 4 of the same part number coils are used for the “primary windings”, 1T1, 1T2, 1SR1 and 1SR2; similarly, 2 identical coils are used for the secondary of 1T (1T3) and control coil for the saturable reactor (SR) 1SR3. There’s no way of telling from the diagram if the cores for 1T and 1SR are the same. From what you say below, it sounds like it might make sense to use the same core design for both 1T and 1SR?

    If the coils on 1T were all on the same core and connected as shown,the two primary fields would cancel and it would just be a resistor.

    Given that the “primary” and “secondary” coils of 1T and 1SR are the same, would you expect the voltage across 1T’s primary to be about 50% of the bus voltage (110VAC bus) when the control current to 1SR3 is near zero and near 100% when the control current is at maximum? (I guess the armature current would never be zero, would it—what sort of minimum current would you expect given a 2HP motor idling a minimum speed?)

    Saturable reactors are current sources. Assuming it is within range, the output voltage will rise until it pushes the current it wants to carry through the load. That current will be the DC current times the turns ratio, with variations from core excitation subtracted, etc. The ratio is typically good in the mid range and drops off on each end. That is why minimizing leakage reactance is important. At the top of the range you can end up pumping huge currents through the control winding with little increase in output. In a good system, which I assume this is, the current fed to 1T, and consequently the voltage developed across the primary will be low. SR should be able to swing to a very low voltage drop, putting almost full line voltage on 1T.

    On the other hand, the SRs in Pete’s 3-phase machine (a total of 6) are each wound on a separate core and mounted in pairs. 1SR and 2SR are mounted together. Pete tells me that there are 6 wires going to each SR coil; 2 are of a heavier gauge than the other 4; apparently there are three windings on each core. The cores for the SR’s are rectangular, with a hollow center. The coils are wound on one side of the core and fill the center of the core. Pairs of SRs are mounted with the windings to the outside with no contact between the cores, apart from the mounting bars.

    "Magnetic Amplifiers Theory and Application" by Sidney Pratt says the separated cores are used to reduce mutual induction.

    The construction of Pete’s SRs appears to be similar to the current transformer on the main panel, 1CT. It has the same core geometry as the SRs. It appears that 1CT’s cores are made up of U-shaped laminations that are slid through the center of the coil, with adjacent laminations coming in from opposite ends to form the rectangular core. Pete says he can’t tell if the SR cores are built the same way, but I would think it’s likely; what do you think?

    That is what you would expect. Usually the Us and Is are all on one end in chokes that need an air gap. Transformers need the best magnetic path possible, so they are overlapped, giving a large area for the magnetic lines to cross between laminations.

    Here’s a couple of photos that show 1CT:
    [INDENT]TB3(?) and 1CT (The photo is rotated 90 degrees; the TB should be vertical. 1CT has a blue tag.)

    I only see two wires coming from 1CT. Are you sure it isn't an inductor (choke)?






    Maybe that would explain why there are 3 pairs of big SRs Pete’s machine instead of 3 individual SRs. If the SR’s in each pair were wired back to back, would the asymmetries cancel?

    I would think so.


    So, if we have a 4MA, there should also be 1MA, 2MA and 3MA somewhere, but those labels don’t exist. I’m guessing that 1MA consists of 1SR and 2SR; 2MA is 3SR and 4SR; and so forth. I assume that the SR suffix means “saturable reactor” and MA means “magnetic amplifier”. Pete tells me that a cursory examination suggests that 4MA is part of the field circuit and 1SR through 6SR part of the armature circuit. 4MA seems to connect to the smaller rectifiers, located above the big SRs. The big SRs connect to the bank of large rectifiers, via three bundles of wire that pass through the bulkhead between the SRs and the rectifiers.

    The field control implies that it uses field weakening. 4MA could be part of that circuit.

    I would like to hear more about that. Could you comment on the “magnetic amplifier” shown in this military tech manual (page 45)? They discuss saturable reactors first (page 43), the move on to the “magnetic amplifier”:
    Cal
    The magnetic amplifier shown is a basic one. The bias is to get the reactors up in the mid range where the slope is steepest. The book mentioned above says that the main difference between saturable reactors and mag amps is the use of multiple control windings which sometimes, but not always, are in a feedback loop to increase gain.

    As to the action of the IR comp circuit, normally the armature voltage is increased as more current is drawn. The idea is that armature resistance introduces a voltage drop subtracted from the voltage driving the motor. Loses in the supply also have the same effect. The compensating circuit makes up for the loses. If you had a motor with no armature resistance and a supply that had no droop, you would have a constant speed under varying loads. Because SR is a current controller, the resistor across 1T acts as a shunt, changing the overall gain of the system by bypassing some of the current from SR.

    Bill

  20. #40
    Join Date
    Sep 2002
    Country
    UNITED STATES
    State/Province
    Arizona
    Posts
    3,631
    Post Thanks / Like
    Likes (Given)
    828
    Likes (Received)
    484

    Default

    Quote Originally Posted by 9100 View Post
    Quote Originally Posted by Cal Haines View Post
    1T and 1SR are each drawn as 3 coils on common cores (one core for 1T and another for 1SR). The “secondary” coils are drawn with fewer turns than the “primary” coils, as you surmise. The “Electrical Stock List” in the upper left-hand corner of the diagram lists the “Transformer Coils” used in 1T and 1SR. A total of 4 of the same part number coils are used for the “primary windings”, 1T1, 1T2, 1SR1 and 1SR2; similarly, 2 identical coils are used for the secondary of 1T (1T3) and control coil for the saturable reactor (SR) 1SR3. There’s no way of telling from the diagram if the cores for 1T and 1SR are the same. From what you say below, it sounds like it might make sense to use the same core design for both 1T and 1SR?

    If the coils on 1T were all on the same core and connected as shown,the two primary fields would cancel and it would just be a resistor.
    I don’t understand that. The primary windings of 1T are wired so that the flux will travel in the same direction, with the fields reinforcing each other. This is how a transformer with a dual-voltage primary is configured in the low voltage mode, is it not? The “primary” windings for 1SR are wired with the polarities opposed, so that the flux of each opposes the other and the fields cancelling, but that’s normal for a saturable reactor, is it not?

    Here’s the armature section of the diagram again, for your reference. The left half of the full diagram (not shown here) does show all three windings of 1T on a common core. I marked the polarities of the windings using the transformer diagrams and terminal numbers shown on the left half of the full diagram (section attached).


    What am I missing?

    Quote Originally Posted by 9100 View Post
    I only see two wires coming from 1CT. Are you sure it isn't an inductor (choke)?
    I can see at least one more wire hiding behind that two in front in some of the photos. The tag on 1CT says “Current Transformer” and lists the ratio as 12:1.

    Cal
    Attached Thumbnails Attached Thumbnails hendy-9in-wiring-diagram-b-c-drawing-deda-5-0-1t-1sr.jpg  


Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •