Please find pdf version here.
This document illustrates how to wind the transformer used to obtain house-keeping power supply voltages on typical small circuits. Such transformers have been used on PCBs .
This transformer is used in a blocking-oscillator configuration to generate the relevant low voltages required to run chips on a typical PCB design. Power levels up to a couple of watts can be supplied by this transformer. The input and output sections/windings are completely isolated with respect to each other. In this particular example the transformer output generates a +5V, +9V, -9V and another isolated (with respect to the these three supplies) power supply of +5V. Typical currents drawn from the supply: +5V@500mA, +9V@100mA, -9V@100mA, +5V@100mA. So the total power provided is approximately 5Watts.
The following materials are required to wind this transformer:
1) EE 25x7 Ferrite core
2) Bobbin (horizontal : implying the core lies horizontally) for this core
3) Wire: 35 SWG (Standard Wire Gauge) enameled wire, or dual coated wire
4) Cello-tape (10-12mm width) or Taco Tape
5) PVC insulating tape for SMPS design (5mm wide roll)
6) Teflon tape – used by plumbers – available at any hardware store
7) Fevikwik or any equivalent cyano-acrilite adhesive
8) Glue stick pieces – for securing the winding in certain steps
9) Tools: Thin triangular file, knife, soldering iron/solder wire.
Ferrite Core and Bobbin for use in housekeeping supply
Material required: Tapes and Glue
The following diagram illustrates the winding details of the transformer. The most important thing to follow is the dot-convention. The diagram along-side shows the winding details for our house-keeping transformer. The view is from the top. That is as you see it from top of the PCB. The pins are on the other side of this view. All the three primary side windings (mains, 230V side) are connected to pins on one side of the bobbin, and all the low voltage secondary side windings are on the other side. This provides for an extra degree of safety to isolate the primary and secondary side of the transformer windings with respect to PCB connections and tracks.
Between each winding, electrical tapes of various kinds are applied, to provide adequate insulation, and prevent shorting of turns. Summary of the number of turns in each winding:
1) NP : 100T of 35 SWG wire on layer 1 - 2 or 35, between pins 8 (dot) and 9
2) NP2: 5T of 35 SWG wire on layer 3, between pins 9 and 10 (dot)
3) NP1: 5T of 35 SWG wire on layer 3, between pins 6 (dot) and 7
4) NS1: 8T of 35 SWG wire on layer 3, between pins 4(dot) and 5
5) NS2: 4T of 35 SWG wire on layer 3, between pins 2 and 3(dot)
6) NS3: 8T of 35 SWG wire on layer 3, between pins 1 and 2(dot)
The most important thing about these transformer designs is to strictly adhere to the dot-convention. Dots indicate phase relationship (same or opposite polarity) of different windings. Two windings that have dots on one of their terminals will have the same voltage polarity with respect to the other terminal, and any time.
Physically, this means that both windings, say starting from the ‘dot’ end, will have the same sense of turning around the core (say clockwise for both or anticlockwise for both). Both turn the same way, and this is the most important thing to remember in dot conventions. If they turn the opposite way, then the phase difference will be 180 degrees (i.e. of opposite polarity).
To avoid confusion during winding, it is best to mark small dots on the plastic bobbin using the hot tip of your soldering iron. It is also good to mark pin 1 with a ‘1’ mark, so that all the other pins may be located unambiguously (follow the same convention used in IC pin numbering). In the images below the dots are marked. I did not get to mark the ‘1’ at this stage, but did it at the end. It is better to mark at the start itself.
The following procedure illustrates the step-by-step construction of this transformer. Most of the steps are indicated with images, so that there is no confusion about the steps.
Step1: Wind NP
Start NP, the primary side winding. We start by removing insulation from the end of the enameled wire (35 SWG), say 15mm and then tinning that end. Then solder it to pin8 of the bobbin.
This winding will terminate on pin9 (dot). In the steps we followed, we started the winding sense going counter clockwise (from our view in this photograph).
The first few turns: This shows how NP is wound. The wire must be tightly wound (without stretching or bending and kinking of any kind) around the bobbin. Any gap between the turns must be avoided. Try to keep the gaps minimized by pushing the turns towards the side the winding star
Dimensionally the bobbin can take up to 56-57 turns. Our target is to put 55 turns on this first layer.
Once the first layer is done, continue to wind on layer two, i.e. above layer 1, until the winding reaches the starting end once again. Typically, if the winding is done with care (remember we are doing this winding manually), one should be able to accommodate 55 turns on each of the layers. The two layers together will therefore provide the full 110 turns which are required for NP.
The photograph here shows the two layers of NP done. The wire terminates on pin9 (dot), which is not yet soldered.
This wire will continue to make the next winding to pin10. But first the surface of this winding must be insulated with tapes. Notice a few small errors during the winding led to a little anomaly close to the center of the winding.
Step2: Insulate NP
The Teflon tape being soft, makes a nice insulating layer, just above NP. The surface becomes smooth when the softer and fragile Teflon tape insulation is enhanced with cello-tape for mechanical strength. The other windings are may be put on top of this insulating layer. Now that NP is secured well in the bobbin, one can remove insulation from the enameled wire and solder it to pin9 (dot) of the bobbin.
Step 3: Wind and insulate NP2
The NP2 winding consists of only 5 turns of wire (35 SWG). For winding NP2, pin10 is the dot end. So, when starting from pin9, one winds the wire going in the direction of pin8 (i.e. counter-clockwise, as was done with NP). The winding terminates on pin10, which is the dot end.
It is important to make sure that the PVC tape insulation is also provided between the return wire (going to pin10 and the already wound 5 turns of wires of NP2, starting from pin9. This process will be true for all such windings.
Step 4: Wind and insulate NP1
Now all three primary side windings are done (pins 6-10). As the photograph shows, these windings are all well insulated from any other winding that may be created for the primary side (pins 1-5). In this application, the secondary (low voltage) windings have very few turns, and they could all get accommodated on the remaining half of the layer3 space. These secondary windings must also be well insulated with respect to each other.
Step5: Wind NS2
Notice that pin3 is dot end with respect to pin2 for winding NS2.
The important thing about this transformer is that the dot convention may be followed separately for all the primary windings or all the secondary windings. Between all the secondary windings, NS1, NS2 and NS3, the dot convention must be followed.
So here, we actually have the winding’s direction sense reversed as compared to the primary. Here we have starting at dot end go counter-clockwise (remember the bobbin is now seen from the other side!).
Step6: Wind and insulate NS3
NS3 was wound again making sure that the cross-over of its end going to pin1 is insulated, using PVC tape, from its 8 turn winding below it.
Cover this winding with PVC tape so that the final winding NS1 can be done on top of it.
Pin 2 is dot end, for this winding between pin1 and pin2. 8 turns, closely spaced, and wound closer to the secondary side of the bobbin, ensures decent functional properties.
Step 7: Wind and insulate NS1
A final look at the dots on primary and secondary sides: