How to set reference for a sine wave with dc supply

[anishkgt]

That voltage is VPP at the secondary with an RMS of 35vac at the primary. The actual voltage at the secondary is VPP 8VA and RMS 2.6VAC. Can you please explain your circuit on the same pdf as i've attached, i find it hard to follow the previous.

3.2 vac is not much to start a high current needed to spot weld.

But why would you consider this voltage as something to start a spot weld. Spot welding has low voltage and High current. Here the voltage is 2.6V RMS.

[C]

I did not catch that you were driving the transformer with 35 VAC.

Using 35 VAC is a great idea during test.

To properly design something you need facts.

Need to know the power consumed by transformer with nothing connected to secondary_______

Using your 35 VAC source

Put a resistor in series with primary winding.

Source voltage = 35 Vac?
Resistor Voltage = ____
Transformer primary voltage = ____
Resistor value = ____

You will need to use resistor that handle the power dissipation.

With above values, can compute now load power use.

I do not want you to harm your 35 VAC source so for the following select a low resistor value that will not harm you 35 VAC if connected across this supply.
Use this resistor as above in series with primary.
Short the secondary of transformer.
Do above measurements again for this state.

while I look at last attachment.
 

[C]

In SWP.pdf you have a box labeled Zero Cross Detect.

The source to this box is 9 Vac

In the data sheet for this AC opt-coulpler you will find some facts or graphs of the led's voltage & current.
You will also find the max safe current for this LED's
Peak voltage  12.72 = 9 Vac x 1.414
Resistor voltage = supply peak - LED forward voltage.

Resistor value = Resistor voltage / <Peek LED current

Now it is a good idea to stay away from limits and a higher value resistor will allow higher peak voltage from source.

Now looking at data sheet again you should find some facts about how the output transistor changes due to current through the LED.
From this you can compute what source input voltage will cause output to change.
Now there is no real need to use such high currents through the LED. Increasing series resistor value will lower LED current and cause a small change in turn-on voltage.

You show an AC input opto-coupler. So from 0 to X volts the output is off and above this voltage output is on.
With AC input this results in output being on except a short time around 0 Vac. Two on's per cycle of AC

[anishkgt]

They work as of now. Just need some help with the contact sense part of the design. Did you manage to ad your version into it.

[C]

To select a value for my R2 need facts about the microwave oven transformer.

Asked you to do some tests using your 35 Vac supply to get these facts two posts back.

[Zero999]

Have you considered a current transformer?

A precision rectifier circuit and comparator could then be used to monitor the current and it has the added bonus of being able to adjust the threshold, rather it being fixed, as in the case of an opto-coupler.

[C]

Hero999
This chain of posts is about changing from a foot switch to trigger spot weld to a contact sense trigger.

Think when it's all done, anishkgt will find that having spot weld contacts spring loaded and sensing pressure/displacement will do a better job.

Yes a current transformer could do sensing part, but first need current limited output to get contact sense with out doing spot weld.

[anishkgt]

So i just checked the optocoupler with the input voltage from the secondary and they seem to be working. I was thinking of a low pass filter to cut off at 100Hz comprising of a resistor of 15.9k and a ceramic cap 0.1uf.

Can i use 5x1k and a 10k resistor all in series or should it be one 15k resistor ? About the TRIAC short pulse of 50ms Max should be ok as long as they are not continuous.

Attached schematic with a current sensor and a fuse.

[C]

If the optocoupler  output is an input to a micro controller you can easily handle the pulses in software.

In addition with a little better software, you can get a better check as if it is not a series of pulses then something is wrong.

Resistors in series is same as total series resistance. With proper layout series resistors will allow higher voltage across the chain.

A TRIAC is for AC, Easy to turn on and stays on until less then holding current.
Not normal to try to turn off TRIAC.

 

[anishkgt]

I have two things to decide on, should i go with a TRIAC or an SCR. TRIACs have lower current ratings than SCRs but conduct both ways unlike an SCR which only conduct one way.

If i were to go with SCRs i would need two in a back to back configuration.

[C]

First look at primary side vs secondary side.
1 to 1.5 volts drop while on is a small % of line voltage compared to secondary side voltage.

In primary side you have smaller currents and less heat generated in Trac or SCR
You also have much less power loss from these.

To get a good spot weld, you need to keep all secondary loses low and have a very low resistance path.

So keep it simple, on primary side the trac or scr will just need to survive primary side current while secondary is shorted, much less.

You stated a microcontroller, when done correctly the analog part can be very simple.

First step is better data about your microwave oven transformer.

With out this data, you are just guessing on currents. The secondary short current during spot weld could blow the house breaker.

The power line is dangerous so use your 35 Vac supply to get some facts.

Just need some simple tests with good facts.
As I asked a few posts ago
A series resistor to primary winding with you measuring some Vac values.
Source voltage, Resistor voltage, primary voltage, secondary voltage.

Do two sets of measurements for each series resistor. One with secondary open, One with secondary shorted.

Start with a resistor with higher resistance as too small could harm your 35 Vac supply.
Need to see some change between open & short for all but Source voltage.

Source voltage should be close to 35 Vac but will change due to load.

Even with this data it will still be some of a guess as to what will happen at 240 Vac.

[anishkgt]

A series resistor to primary winding with you measuring some Vac values.
Source voltage, Resistor voltage, primary voltage, secondary voltage.

Do two sets of measurements for each series resistor. One with secondary open, One with secondary shorted.

Start with a resistor with higher resistance as too small could harm your 35 Vac supply.
Need to see some change between open & short for all but Source voltage.

The only resistor i have with a decent current rating is 1K 1W, Would that be ok ?
Well values are as follows

AC voltage: 34v
voltage at Primary: 29.3v
@Resistor (1K 1W): 16.8v
Secondary open : 0.19v
Secondary shorted resistor voltage: 26.3v

[C]

Should grab the shorted values for others due to load change.

[anishkgt]

Secondary shorted resistor voltage: 26.3v

This is the voltage at the resistor when the secondary was shorted.
Voltage at the primary when secondary (with resistor in series)is shorted : 27.4v

[C]

Think you some wrong values.

Source voltage = resistor voltage + primary winding  voltage when you have 1K resistor is series

voltage at Primary: 29.3v
@Resistor (1K 1W): 16.8v

This = 46.1 volts for source voltage.

or meter is lag from transformer

[C]

If your meter gives these values you will need to look at signal with OSope.

The inductance of transformer can cause current lag from a pure resistance circuit.

[anishkgt]

I am not sure if it would ok to check the ac voltage with a scope. The 34v does not come from isolated source but comes from the same line as the PCB is powered. The 34v is just a phase shifted voltage.

I could check secondary side but not sure how to do it at the primary without blowing the scope.

[C]

OK
  might not be a great idea with scope.

Starting over.

To sense a contact between the two spot weld contacts you need to create a current flow or voltage change.

You stated you are using a microcontroller.

I see you have created a Zero-voltage cross circuit.

This way is not nice to power line

Fpr creating spot weld you turn on TRIAC after zero volt cross but near 0. Due to inductance causing current lag, TRIAC will actually drop below holding current after zero volts.

For contact sense you turn on TRIAC a short time before Zero cross. Again off is after Zero cross.
This will create a short voltage spike on secondary side and can show with a primary side current sensor. Open will be one low current value. Contact will be higher current value.

Just so you know, how hot a spot weld gets is heat vs time or here number of 1/2 cycles of AC in this case.
Contact would be too to little heat to do anything but sense contact.

The advantage is no series resistor in primary as contact will use current sensor.

To keep it simple, really should use ADC to measure Current but this will take better software then just contact & no contact.
Creating digital input is not that hard so  software could be just a digital input pin.


So micro will sense Zero cross of line
Will sense primary current with ADC or digital pin
Will control TRAC with driver that does NOT have zero cross switching.

[anishkgt]

I am trying to make sense of what your trying to say but half way through you seem to jumping between ideas and the first part leaves me puzzled.

Just to make it clear the 33vac is actually produced by the microcontroller by switching on the SCRs at 9000ms after the zero cross. I could increase the voltage by switching on the SCRs few seconds after the zero cross, like maybe 7000ms or 8000ms. So you say i put a resistor in series with the primary and determine when electrodes short by measuring the voltage difference ? well what happens when at full voltage ? I may have misunderstood but without a clearer schematic am just guessing.

Mean time I've been thinking at the secondary side. The TRIAC here is the BTA440Z-800BT (http://www.ween-semi.com/documents/BTA440Z-800BT.pdf) The non-repetitive peak on-state current (I-tsm) at 20ms is 400A. So based on that comment in the datasheet i guess this should work but not sure about the heat dissipated and how long the TRIAC would hold up.

[anishkgt]

here

Nice an simple

In your schematic, as i understand it, the resistor in series with the opto-coupler should be high enough drop the voltage to zero when the secondary are shorted and low enough such there is some voltage for the LED in the opto-coupler to work, am i correct ?

[anishkgt]

seems to be simple but just not sure if D8 should be a zener or just a generic Diode ?