Looking for information on Tyco SE032S module

[LinuxHata]

Hello.

A friend of mine found these modules at surplus sales and bought a bunch for me.
While these are on the way, I'd like to get at least some info on them.
Googling brings up only this ebay listing, and as it seems, seller also has no idea on them.
https://www.ebay.com/itm/323812272425

As I can understand from the circuitry, this appears to be a non isolated, high voltage (<400V) flyback boost module.
I'd like to know amp rating at least, based on marking I can guess it will be 32W ?

Answer might be in the flyback transformer model, but I can't find any info on the maker - whenever it is SAE or SRE ?

Any ideas?

[m k]

ModuleS, how many, some test subjects?

Maybe the maker is SE and middle mark is something else.

[lunar]

If you have more than one of these, take one of the transformers apart, count the windings, and measure the diameter of the wire. Compare to other transformers on mouser, digikey, farnells, etc... Diode gone wild has a number of videos reverse engineering switchers, so if you watch a couple of those, that might help.

[LinuxHata]

These were received recently.
This is super weird thingy.
Very complicated circuit, which does not works properly.

First of all, it is inverting, non isolated converter.

UCC2803 is the main IC. It is equipped with some UVLO/OVLO circuits and module only works in range of 5-6 volts. But even then, this "working" is very strange - there's about 16v dc output, which goes to 3V, even with 220 ohm resistor as load.

There is a "SYNC" pin, on which is 4VDC, and whenever you tie it to GND or tie to VIN - there's no chances.

The transformer has 3 windings, primary, secondary and current sense (my guess). Primary is 50uh and secondary is 5mh.
Output inductor has 1mh inductivity and 1UF output capacitor is used.

The amount of components used is truly amazing - there are 23 SOT-23 cases, not counting large number of resistors - for example, they have made current sense shunt for mosfet from 6 pcs of 2.2R 1206 resistors.

[Gyro]

These were received recently.
This is super weird thingy.
Very complicated circuit, which does not works properly.

First of all, it is inverting, non isolated converter.

It's a bit harsh to say that it doesn't work properly. It presumably meets the (unknown) board spec / requirements. It looks like a customer specific board, that was designed and made to order by Tyco, as part of the customer product rather than a catalogue off-the-shelf product. One of the things that was required was clearly an inverting output.

The amount of components used is truly amazing - there are 23 SOT-23 cases, not counting large number of resistors - for example, they have made current sense shunt for mosfet from 6 pcs of 2.2R 1206 resistors.

That's pretty common for a volume manufacturer, generic discrete parts are verging on free (mostly placement cost). Using multiple small paralleled current sense resistors is very common in volume products, power bricks etc, it's cheaper than a sourcing a specific large body 370mR smd resistor and the value easier to tune.

[LinuxHata]

These might be factory rejects - QC failure batch or something like that (I've paid about 25 cents per board). So quite possible this is why they don't work. It has too few pins to mess around or to have any sort of control I'm not aware off.
I'll try to apply PWM to the "sync" pin and see if anything moves.

[squadchannel]

These might be factory rejects - QC failure batch or something like that (I've paid about 25 cents per board). So quite possible this is why they don't work. It has too few pins to mess around or to have any sort of control I'm not aware off.
I'll try to apply PWM to the "sync" pin and see if anything moves.

I think so.

is marked "CASE" on the upper left corner of the board, and i believe it is 1” square size.
it is probably a qc failure 1" brick DC-DC module. there is some silicon-like residue on the edge of the board.

if the manufacturer is TE, it could be Lucent, AT&T, Lineage power, or Cherokee.
the same products were being made by all of these companies at this time.

[LinuxHata]

PCB dimensions are roughly 68x53mm.
"CASE" is next to a pin, which is not connected to anything. Maybe it was meant to connect to metal shield, which goes above the assembly? but why then sticker with serial number on the heatsink?

[LinuxHata]

Here's top view.

Mosfet under the heatsink is IRF630.

The font used to write "SYNC", "Vi and shape of brackets are very similar to these I've seen in NEC/Mitsubishi modules of similar type. In fact, if not "Product of Mexico" I would consider this as Japanese made. Even optocouplers are NEC.

[prints-f-darkness]

Might be a custom part, closest thing I could find was a datasheet for a Tyco SE014S110 (attached), but it does look like a DC-DC converter.

https://web.archive.org/web/20010703011252/http://power.tycoelectronics.com/pdf_dcdc.html

[LinuxHata]

Thanks!

At least we now know that these modules are rated 32W
But as input voltage seems to be 5V, 32W at 5V is almost 7 amps, and input pins do not look like they were made to handle such current.

[CaptDon]

Tyco/Simplex was heavily involved with fire alarm systems. Perhaps that board was used in one of their alarm control panels? I wonder if the 'sync' signal controlled a 'flash' device making all fire system flash devices flash in unison?

[LinuxHata]

Well, I have taken a lot of "flash" alarm systems, and circuitry never was so complex - output voltage is tightly controlled here, there are two optocouplers, TL1431, LM2904 and 4 different SOT-23 components on the hot side. On cold side there are even more SOT-23 components. Also, output inductance is 1mH and wound with very thin wire - definitely no need for high voltage capacitor charging.

Here are close-up photos. Please do not pay attention to red dots - these are my phone case reflections.