# Replacing AWG in transformer's primary coil. How many turns...?

Status
Not open for further replies.

#### J_Nichols

##### Member
I have a small core with only a primary coil (input). This wire has been broken, so I need to replace it.
The problem is I only have another AWG wire, so I need to use the new thicker wire.

I don't know how many turns I need in the primary. How can I calculate it?

The transformer is a part of an air pump. So it has only the primary and has not secondary coil.

#### MaxHeadRoom78

##### Well-Known Member
One way normally to find the turns/volt is to wind on say 10 turns and measure the resultant output, this gives the t/v.
When you say broken how many turns have been taken off already, if any?
Max.

#### ronsimpson

##### Well-Known Member
Most Helpful Member
measure the resultant output
The wire is broken, so can not.

Unwind the wire and count.

#### JimB

##### Super Moderator
Most Helpful Member
The transformer is a part of an air pump. So it has only the primary and has not secondary coil.
If it has only one winding, it is not a conventional transformer.
It could of course be an auto-transformer.

If it is part of an air pump, is it simply a solenoid for moving the pump diaphragm?

JimB

#### J_Nichols

##### Member
One way normally to find the turns/volt is to wind on say 10 turns and measure the resultant output, this gives the t/v.
When you say broken how many turns have been taken off already, if any?
Max.
Where do I should measure the output? because it has only a primary coil. It has no secondary.
I haven't taken any.

#### J_Nichols

##### Member
If it has only one winding, it is not a conventional transformer.
It could of course be an auto-transformer.

If it is part of an air pump, is it simply a solenoid for moving the pump diaphragm?

JimB
It's used to move the diaphragm, of course. It has a u-shaped core and it has a solenoid around one leg of the U.

#### MrAl

##### Well-Known Member
Most Helpful Member
Hi,

One of the equations sometimes called the "Transformer" equation is:
B=E*10^8/(4.44*F*A*N*S)

where
B is the flux density in Gausses,
E is the AC rms voltage in volts,
F is the frequency in Hertz,
A is the core cross sectional area in square centimeters,
N is the number of turns,
S is the stacking factor, which ranges between about 0.8 to about 0.95 depending on if the laminations are interleaved or butt stacked.

For transformers made to run off the line voltage the core material is usually Si steel which has a max B of about 20KGauss, but using a max of about 15KGauss allows for some room at the top so it does not saturate.
The core cross sectional area is measured by multiplying the width of the center leg times the stack height.

Another thing you could do is calculate the required number of turns with a lower voltage, wind the required number of turns, then aopply that lower AC voltage and check the input excitation current to see that it is not too high. You may get some buzzing too because the windings will not be varnished, and if the core is not varnished you'll get more audible buzzing.

#### spec

##### Well-Known Member
Most Helpful Member
I have a small core with only a primary coil (input). This wire has been broken, so I need to replace it.
The problem is I only have another AWG wire, so I need to use the new thicker wire.

I don't know how many turns I need in the primary. How can I calculate it?

The transformer is a part of an air pump. So it has only the primary and has not secondary coil.
Can you tell us more about the solenoid. What is the voltage across the coil. What current does the coil take? Is the solenoid powered from AC or DC? What was the coil resistance?

By the way, if you have just lost a few turns off the coil, it will not matter- just reconnect the existing wire and call it a day.

Alternatively, if you have lost a significant number of turns just connect your new wire to the existing wire, insulate the joint with heat shrink sleeving for example, and wind the missing turns back on.

As long as the new wire is thicker all will be well.

But, if you have lost many turns you will need to be more scientific and the answers to the questions above will help determine the best approach.

spec

#### J_Nichols

##### Member
If it has only one winding, it is not a conventional transformer.
It could of course be an auto-transformer.

If it is part of an air pump, is it simply a solenoid for moving the pump diaphragm?

JimB
Here you have the picture

#### J_Nichols

##### Member
Can you tell us more about the solenoid. What is the voltage across the coil. What current does the coil take? Is the solenoid powered from AC or DC? What was the coil resistance?

By the way, if you have just lost a few turns off the coil, it will not matter- just reconnect the existing wire and call it a day.

Alternatively, if you have lost a significant number of turns just connect your new wire to the existing wire, insulate the joint with heat shrink sleeving for example, and wind the missing turns back on.

As long as the new wire is thicker all will be well.

But, if you have lost many turns you will need to be more scientific and the answers to the questions above will help determine the best approach.

spec
The voltage across the coil is 230 VAC.
The current I'm not sure, but the air pump consumes 1 watt.
I have re-wired and the coil resistance is 18.3K. I have just lost some wire.

#### J_Nichols

##### Member
Hi,

One of the equations sometimes called the "Transformer" equation is:
B=E*10^8/(4.44*F*A*N*S)

where
B is the flux density in Gausses,
E is the AC rms voltage in volts,
F is the frequency in Hertz,
A is the core cross sectional area in square centimeters,
N is the number of turns,
S is the stacking factor, which ranges between about 0.8 to about 0.95 depending on if the laminations are interleaved or butt stacked.

For transformers made to run off the line voltage the core material is usually Si steel which has a max B of about 20KGauss, but using a max of about 15KGauss allows for some room at the top so it does not saturate.
The core cross sectional area is measured by multiplying the width of the center leg times the stack height.

Another thing you could do is calculate the required number of turns with a lower voltage, wind the required number of turns, then aopply that lower AC voltage and check the input excitation current to see that it is not too high. You may get some buzzing too because the windings will not be varnished, and if the core is not varnished you'll get more audible buzzing.
Do you mean that I need to apply a minium amount of amperes to induce gauss in the core?

#### MrAl

##### Well-Known Member
Most Helpful Member
Do you mean that I need to apply a minium amount of amperes to induce gauss in the core?
Hi,

Not sure what would give you that idea because there is no 'amperes' in that formula right?

That just tells you how many turns you need when you have a certain line voltage E and frequency F and with a core with area A and stacking factor S and max flux density B. Most transformers made for line frequencies have max 20000 Gauss but we have to leave some room for line variations so 15000 Gauss would be better.

In a form with area A in square inches we have:
B=(3490998*E)/(A*F*N*S)

and solving that for N we get:
N=(3490998*E)/(A*B*F*S)

and written a little shorter:
N=E*3.49e6/(A*B*F*S)

For example with a core that is 2 inches by 2 inches (center leg) and with frequency 60Hz and voltage 120vac rms and stacking factor 0.95 and using 15000 Gauss for the max flux density we get:
N=122 turns for the primary.

We calculate the turns so that we have enough turns on the primary because if we have too few turns the transformer will draw way too much current due to core saturation. If we have too many turns we have an unnecessarily high primary resistance.

In your pic in post #9 i see part of your core, the "E" part, and what looks like the coil, but i dont see the "I" part of your core. Your core does have an "I" part too right? If not it will never work right.

Last edited:

#### spec

##### Well-Known Member
Most Helpful Member
The voltage across the coil is 230 VAC.
The current I'm not sure, but the air pump consumes 1 watt.
I have re-wired and the coil resistance is 18.3K. I have just lost some wire.
I would suspect that would be fine, so long as you have not lost too many turns.

spec

#### J_Nichols

##### Member
I would suspect that would be fine, so long as you have not lost too many turns.

spec
so, it's sure to replace with a thicker wire and wiring the same number of turns as the old wire?

#### J_Nichols

##### Member
Hi,

Not sure what would give you that idea because there is no 'amperes' in that formula right?

That just tells you how many turns you need when you have a certain line voltage E and frequency F and with a core with area A and stacking factor S and max flux density B. Most transformers made for line frequencies have max 20000 Gauss but we have to leave some room for line variations so 15000 Gauss would be better.

In a form with area A in square inches we have:
B=(3490998*E)/(A*F*N*S)

and solving that for N we get:
N=(3490998*E)/(A*B*F*S)

and written a little shorter:
N=E*3.49e6/(A*B*F*S)

For example with a core that is 2 inches by 2 inches (center leg) and with frequency 60Hz and voltage 120vac rms and stacking factor 0.95 and using 15000 Gauss for the max flux density we get:
N=122 turns for the primary.

We calculate the turns so that we have enough turns on the primary because if we have too few turns the transformer will draw way too much current due to core saturation. If we have too many turns we have an unnecessarily high primary resistance.

In your pic in post #9 i see part of your core, the "E" part, and what looks like the coil, but i dont see the "I" part of your core. Your core does have an "I" part too right? If not it will never work right.
Thanks for the information. It hasn't an I part. Only U part.

#### spec

##### Well-Known Member
Most Helpful Member
so, it's sure to replace with a thicker wire and wiring the same number of turns as the old wire?
If you wound the coil with a thicker wire, it would be unlikely that the solenoid would take excessive current, providing that the new coil had roughly the same number of turns.

But the solenoid may be designed to use the coil resistance to limit the input current and hence the heating of the coil.

Best to add a resistor in series with the new coil to make the total resistance seen by the mains supply the same as the original.

Then gradually reduce the resistance, each time ensuring that the coil does not exceed 70 Deg C.

If the solenoid does not reach 70 deg C or close without a resistor leave the resistor out and job done.

Just one point. The windings on a solenoid are not as critical as on a transformer.

spec

Status
Not open for further replies.

Loading