Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

IN6296 - Any one know about this

Status
Not open for further replies.

Quantised

Member
Hi All

I have a circuit with a pair of IN6296 diodes that appear on the cirduit diagram as zener diodes and in line with 2 off 2.2Kohm resistors. This is tapped of from a 240V supply and excites a 4N35 which I presume to be a way to detect if power is applied.

However when I search for this IN6296 I find no data any where and so if any one knows what it is please let me know.

From the circuit board it seems they got really hot and burned the board so I was thinking to replace them - but with what?

Thanks

Quantised
 
OK I am not your electronics engineer so what does it do. It seems to get really hot so maybe it is not a 96v zener as I hope it would be.
 
Thanks for the link and it explains what it does. Some manufacturers seem to quote 98v min breakdown and others 106 up to 115 but I suppose mine is around 100 as I have 220V input.

If my numbers are correct the whole lot dissipates 1080 mW which should be OK but if one of the IN6296 has blown it will get pretty hot with over 6000mW dissipated.

Does a blown IN go open circuit or direct short?? If it went open circuit the mains detect would surely fail to show mains is available?
 
Last edited:
What you can do is to look at the voltage across each of the diodes. If you have a scope, you can use A-B or a differential measurement.

Note the datasheet has min max values instead of say a nominal +-X%. They are just better numbers to design by.

What may have happened is the 2.2K resistors changed value and got lower in value. Those resistors should be metal oxide if possible. If you replace them, elevate the diodes and the resistors above the board slightly. So, those Transorbs are basically being used as a Zener regulator.

The circuit is also a little wierd. You could likely do it better and with less power dissipation by using a series RC network.
 
Last edited:
The circuit is on a charger control board and it is circa 1985 vintage so trying to keep it going a while longer.

What did you have in mind with an RC network?? I would be grateful to see an outline!

So these components have been working for nearly 30 years and you think they may degrade?

I have the full circuit diagram for the board as it came with the manual (companies do not do that these days so much) and it looks like all it does is detect the mains voltage and turns on the optocoupler which in turn signals the rest of the circuit that power is available.
 
There are a couple of options. One is based on this IC: **broken link removed** Take a look at the datasheet.

Another is modifying this very cheap AC detector to detect 240 VAC. http://www.gravitech.us/buacinbo.html Refer to the user's manual.

If you take the 100K resistor and the 0.1 uf cap, and plug them into here: **broken link removed** with a L value of zero, you get about 26K of capacitive reluctance. Tke ~170 V and divide by ~26K and you get about 6 mA of current through he optocoupler. 170 is aprox 120*1.414. If you half the capacitance, the circuit will work on 240 V. But, you would also have to check the voltage rating. Putting the caps in series would effectively double the voltage rating for the combination.

If you take a look at figure 1, here: http://www.datasheetcatalog.org/datasheet/SGSThomsonMicroelectronics/mXyzyvt.pdf you can see how that would work.

My brain isn't quite working right at the moment (migraine), but it should be food for thought anyway.

In the datasheet, the 240 is basically full wave rectified to give 300 VDC and for 120 V input, the voltage is doubled, so 120 VAC results in 300 VDC as well. I'll have to stare at the stuff some more to actually get component values.

i.e. Change the cap to 0.05 uf or put two 0.1 uf Caps in series. The caps would have to be an X or Y safety cap. Not sure which is more appropriate.

By using a series combination of R and C, you are using the capacitive reactance to limit the current, so there is less loss.

These sorts of modules are available too: **broken link removed**

Carbon resistors can change value. Metal oxide resistors tend to open when they get stressed. Carbon composition tends to go to a higher value. Metal film resistors puddle and sometimes get lower in value.
 
Last edited:
Thanks for the link and it explains what it does. Some manufacturers seem to quote 98v min breakdown and others 106 up to 115 but I suppose mine is around 100 as I have 220V input.

If my numbers are correct the whole lot dissipates 1080 mW which should be OK but if one of the IN6296 has blown it will get pretty hot with over 6000mW dissipated.

Does a blown IN go open circuit or direct short?? If it went open circuit the mains detect would surely fail to show mains is available?

I missed this one. Diodes will normally short. The zener breakdown is non-destructive if the current is limited. This is basically the same as the PRV (Peak reverse voltage) of a normal diode. Measuring the voltages across the devices in series would be useful. The 2 resistors, the transorbs and the diodes. Even the ones in the bridge and the LED. The cold transorb would be the bad one.

The LED's in an optocoupler can take a fair amount of abuse. e.g. 10 mA nominal and 70 mA and then a peak rating.
 
It took me a while to go through all the data sheets you posted and I am getting a handle on how it all works. The 4N35 that is on my board seems similar to the MID400 except with different input ma to the LED and the MID has its own internal reference level but I cant really see what is.

One of the reasons the circuit has 2 off 2.2K and 2 off IN6296 is that there is a jumper across a resistor/TVS pair that is removed when 240V input is used and put in when we run on 120V. Also if the voltage is below the TVS voltage the circuit wont fire and so show mains is too low. I thought about changing to just resistors but then this key detection TVS would not be doing its job.

So I have bit the bullet and ordered from RS some new diodes, resistors, some new triacs as the old ones blow about once a year when on 210V (should be 230V) as the mains is a bit erratic.

Also think I will put some petals on the lead wires to help cooling.

Also ordered an MID400 and bits to suit it so I can make up a piggy board to replace the TVS if it is a failure.

20bucks of parts and 85bucks to send it hurts but not as much as going through the RS-online catalogue using a 2g connection,

Am going to measure the voltages around the circuit and see what is happening if I can get a probe to the right spots.

Q
 
Last edited:
I measured the volts using a DC meter and the attached diagram shows the results, the extra wires are to show the clip locations.

Seems they are working fine.

But as I moved the case (it has to be open) I got a nice tingle so assume it was earthed till I moved it and then it floated to some 110V AC or similar. Now it does not work at all so assume something gor fried when the case moved. Nothing was shorted inside with the probes so confused and concerned.

I have the full schematic that I scanned but it is quite large, would it be any help (I hope so!!!)
 
Here is the circuit diagram for this board.

Usually the system buzzes a little - the triac I assume, but now no buzzing and no charging.

Any ideas.
 
You did good. I think just elevating the resistors and the diodes above the PCB would be fine. Those 2.2K really should be metal oxide.

The IC solution probably won't work. It's not detecting the AC line directly.

I don't see a snubber across the triac.
 
Last edited:
I missed this one. Diodes will normally short. The zener breakdown is non-destructive if the current is limited. This is basically the same as the PRV (Peak reverse voltage) of a normal diode. ......................................
Depends upon your definition of "basically". ;)

The zener breakdown voltage is specified with a tolerance.

The PRV is a minimum breakdown voltage. There's no upper limit. It could break down at a voltage much higher than its rating.
 
True, but in a Transorb it's a controlled parameter and the parameter has a tolerance. There are similarities and differences between PRV and Zener/Avalanche breakdown.
 
Any idea why the triac keeps blowing - the original MAC223 was replaced with a BTA26-600 as they had near identical specs. now I loaded a new one yesterday and it has blown again??

It says it is snubberless but what is snubber anyway and do I need it?

As far as I can see the only thing connected to the case is the T03 insulated BTA26 which now seems to be uninsulated and an LM317 with insulation strip and washer.

Any idea what to retrofit to make it work safe as the last BTA26 lasted a good year.

Thanks

Q
 
I don't see a snubber across the triac.

I have researched snubbers and it seems my new triac is a normal design and not snubberless (needs a W at the end of the product code).

But a snubber is an RC networkk across the triac and there is a 430ohm and 2 off 0.22uF across the triac, is this not a snubber?

Any comments much appreciated.

Q
 
Yep, drawn funny. The 0.22's are in parallel.

I do agree that the substitute that you picked looks like a really good match.

Look for loose connections everywhere.
 
SO do I need a snubberless version or do I need a normal snubber version of the triac?

Will the snubber upset the snubberless triac as this is what is winging its way by DHL.

I have to go with any recomends as it will be a long time before I understand triacs, trigger circuits and phase delays etc........
 
If I can only get 6A or 10 A triacs can I put these in parrellel to get get 25A rating? Say 2 off 10A units might work for a while till the shipment gets here, in 2 weeks time.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top