240V small PIR - any chance I can extend delay time?

Thread Starter

advarp

Joined Jan 12, 2018
57
I am repairing several Arlec sensor coach lamps -old style made properly with glass and metal. I am using LED lamps, only 8W or so power.

I have bought this small 240V PIR off Ebay. It looks a bit flimsy but it's $4 compared to $18 for the relay ones. Considering I have to repair around 6 lamps saving could be substantial.

The main problem with this PIR device is that the delay time is too short and not adjustable currently.

My question is: can i change the delay using e.g a larger capacitor and / or larger / smaller resistors?

Problem is, I have no idea how it works and most components are SMD.

I am posting a few photos, if you need other angles let me know. I could not get the wires out of the way yet as the connector is still on and does not fit through the hole.

Caps are 22uF 25V. Larger caps are not a problem space wise. The circuit is very small - this is good because the original lamps had the PIR and LDR separate from the main circuit. I cloud fit the whole thing in the small PIR sensor case.

I do not need to have controls outside the lamp, just set a time of say 7-8 minutes. It is already working only in the dark.

Any help would be great, thanks,.

upload_2018-2-28_10-31-43.jpeg

upload_2018-2-28_10-32-8.jpeg

upload_2018-2-28_10-33-29.jpeg

EDIT: I found: M7: rectifier; MB10: bridge rectifier; 7130: MOSFET transistor; C106MS Sensitive gate silicon controller rectifier (? what?)

EDIT2: Ha, there seems to be an adjustable version on AE - and see this Russian site with the schematics also... Question is, can i extend the 4 minutes... hm. Might buy one off AliExpress anyway.

EDIT3: The Russian discussion is very interesting - use Google translate... I might get one of AE but I'd like to extend the delay time from 4mins to 8-10 mins...
 
Last edited:

ebp

Joined Feb 8, 2018
2,332
I'd be willing to bet (but not much) that the cap near the brown wire is a timing capacitor. The other one probably is in the power supply for the circuit.
 

Thread Starter

advarp

Joined Jan 12, 2018
57
upload_2018-2-28_11-5-13.jpeg
Off the Russian site (bear in mind this is the one with adjustable sens and time). Seems they use resistors for timings

Question is, for my device, which cap to play with as both are 22uF 25VDC lol
 
Last edited:

ebp

Joined Feb 8, 2018
2,332
Again, I'm guessing.

Both caps are probably the same value just to simplify the bill of materials.
Timing can be done with a capacitor and a current source from some semiconductor device or with a cap and a resistor. The latter is far more common. Within reason, either R or C or both can be increased to increase the time period.

I could suggest a test to identify the timing cap but it requires care and carries a risk of injury because the circuit is connected directly to AC mains and will have to be tested "live" for what I have in mind. I'll make the suggestion if you want to contact a moderator and get clearance. Perhaps someone who has been around AAC longer than I will jump in.
 

Thread Starter

advarp

Joined Jan 12, 2018
57
Again, I'm guessing.
I'll make the suggestion if you want to contact a moderator and get clearance. Perhaps someone who has been around AAC longer than I will jump in.
Lol I'm an electrician and worked with electronics for 30 years - well at hobby level only...

You mean add a cap in parallel with the existing (try both in turn) ? Of course I won't do it when the circuit is live - hehe.

Anyway I'll look at the moderator stuff, if it takes a lot of work I'll just try some caps, if it burns out oh well
 

Tonyr1084

Joined Sep 24, 2015
7,853
I think I agree with the experiment advice. Remove the 22µF 25V caps and replace them with 47µF 25V caps. Not sure if the 47 comes in 25 V but whatever you do DON'T go below the 25 V rating. Likely the caps will be a little larger. But don't put caps parallel. Remove one and replace with larger. Test and see if you like the results. If it runs too long for your liking I think you can find a 33µF 25V rated cap. But this is all experimental. And it's not likely you'll hurt anything.

I'd suggest you replace one cap first and see if that solves the problem. If not - put it back the way it was and then change the other cap. If that changes the dwell time then you've found which cap it is that controls (in part) the dwell timing.

In adjustable PIR's they use either a set of resistors that are switched into or out of the timing circuit. Some have variable potentiometers that allow an infinite adjustment between min and max timing. Seems yours doesn't have any adjustment capabilities, so if you can identify which resistor(s) controls the timing then it's probably easier to change the capacitors. Keep in mind the higher the rating you go the larger the capacitor will be. You say size is not an issue, however, a 3 farad 2.6 volt cap is larger than a child's hand. That's not gonna fit. Of course that's an extreme example, but the point is there IS a size limitation. Changing BOTH caps may eat up valuable space, so change only the one that needs to be changed. It's easier than identifying the resistor and a damned sight easier to change than an SMD resistor. Of course, the resistor IS the proper way to adjust it. But changing the cap will work too.
 

Thread Starter

advarp

Joined Jan 12, 2018
57
Thanks for the advice, will try that. No, I don't intend to use a 3 Farad Cap lol - a child's head might be cheaper than the 3 Farad anyway ? hehe. The idea is that I can have the cap away from the board of course, 47uF should not be that big in 25V.

I purchased the adjustable one also from Ali Express - only $3, but on that one also I'd like to adjust the time (I think is has the same 25uF 25V but uses a series of resistors as per above photos from the Russian site, so there maybe change the resistor). AE takes a while to arrive though so in the meantime I might play with he fixed one.
 

Tonyr1084

Joined Sep 24, 2015
7,853
Well, all you have to do is follow the traces to the first resistor. Remove the resistor and put in its place a couple wires so you can add a potentiometer. But be sure to keep SOME resistance on it so that if you turn the resistance all the way down to zero ohms you won't blow the board. Since we don't know what that resistor is (the one you want to change) we can't make a good suggestion. Well, I can't anyway. Someone else might be able to. It's probably going to be a resistor on the positive lead of the cap. Follow that trace and you'll find the resistor you want to remove.

Lets make up some numbers: Suppose the resistor is 1KΩ along with the cap gives you 4 minutes (as you said), then a 500Ω resistor should give you 2 minutes (remember, I'm making stuff up here). So with a 5KΩ pot in series with the 500Ω resistor will give you no less than 2 minutes but potentially as long as - well - very long. You just want to be sure to keep some resistance in there so that when you turn the pot all the way to one side you don't blow out the unit.
 
Last edited:

Thread Starter

advarp

Joined Jan 12, 2018
57
Well the SMD resistors may actually have the values printed on them but I have to wait until the AE unit arrives. The fixed timing one does have a resistor, I will try to read the value. soldering wires in place of an SMD will be a PITA.

Actually I can see an SMD component with '101' on it. That would be a cap, right? Resistors may only have colours or nothing on them?

I do wonder generally how reliable this unit is, might test it a bit anyway - no relay.

I repaired a lamp already using a different unit with separate sensor and LDR and using a relay for mains, I am sure it is more reliable but it's $14 or so per unit...As usual, $ vs quality I guess...
 

ebp

Joined Feb 8, 2018
2,332
I guess if the mods don't like what we're up to they'll put us rockers down.

SMD caps except for electrolytic types are almost never marked. Ceramic caps are generally shades running from white to tan. Resistors come in assorted colors by virtue of the fact the tops are sometimes painted. The trend is away from marking them in any fashion. If it is black, it's probably a resistor (smd resistors are made with the element up; the body is white ceramic).
The part between the two electrolytic caps marked 101 is a 100 ohm (10 x 10^1) resistor
The tan part near the bottom electrolytic in your first photo is a ceramic capacitor - unmarked.
The part marked C106MS is an SCR (that series goes back decades, originally from GE, I think). The circuit use a bridge rectifier and an SCR instead of a triac, which is ... interesting.

I would tack some suitable wires to the electrolytic cap pads then monitor the voltage across them with a meter during a timing cycle. A meter with a bar graph would be helpful. The timing cap will have a voltage that changes significantly from start to end of timing. There is a reasonable chance the timing cap has nearly zero volts across it when sitting idle (light off, no motion).
Evey point must be regarded as being connected directly to line HOT, safety-wise.

The through-hole caps are far easier to fiddle with and there is (likely) zero reason not to change the timing capacitor. It will cost a few cents vs. a fraction of a cent for a resistor, but will be much easier to pull and replace. Do what you suggested and bridge another across it for testing. There is a possibility that one cap sets two time delays - one for turn-on to reject bats & bugs and one for ON time, but I doubt it. It makes the active circuitry messier.
Electrolytic capacitors that have not had voltage applied for a long time can be a bit "leaky" - very small DC current flow. This normally goes away after several minutes of being powered. While it "reforms" the time delay may be a bit longer.
 

Tonyr1084

Joined Sep 24, 2015
7,853
SMD caps except for electrolytic types are almost never marked.
I've quoted that statement because it's absolutely true. Your component marked 101 is a resistor. SMD components are too small to have color codes on them, and the numbers are often more easily read. 101 simply means the value has a 1 in the first place, a 0 in the second place, and the third "one" represents the number of zeros that follow. For instance, a 100 marking means a 10 ohm resistor because after the 1 and the 0 there are no more zeros to be added. A 101 means there's a 1 and a 0 and one more 0 after, thus, 100 ohms. A 473 means a 4, a 7 and 000 (three zero's) Thus, a 473 is a 47KΩ resistor.

Yes, as ebp said, SMD components are difficult to work on. The smaller they are the harder they are. Some of us have a lot of experience with them, and it takes a lot of experience to be able to do a good job. And as ebp said tacking wires on isn't going to be the easiest route. However, I think having a variable resistance in its place will give you the adjustability you're looking for. But if you only want a longer time then yes, again, I agree with ebp, changing the cap is FAR FAR easier. And the expense is virtually nil.

The important thing to keep in mind is that the capacitor you put in place MUST NOT be rated for a lower voltage. If you replace a 25 volt capacitor then you must replace it with one that is at least 25 volts. Not lower. A larger capacitor will take up more space. I would expect that a 47µF cap will be more than twice the size of a 22µF cap. One might be manufactured to be twice as long. Another might be 50% larger diameter. (yes, 50%. has to do with volume) Still others may be some combination of longer and wider. Space becomes a consideration and sometimes is the trade off between adjustability and sufficient functionality. If you want a longer time always then change the cap. If you want to be able to adjust it, and I have adjustable PIR's too, then you go with either a potentiometer or a switch that can select between a number of resistors. Honestly, once set I never find that I have to change the duration of my PIR's. Once they're set they are usually left alone. I also like to mount them up high enough so that they're not easily tampered with. My neighbor has one that has been tampered with several times. Mostly the light bulbs have been unscrewed because the trespassers have wanted to keep the lights off. And she HAS had attempted, but failed break in's.

Once you find the timing you want, just build it that way. Forget about adjustability, it's a nice luxury, but a not so necessary one.
 
Last edited:

Thread Starter

advarp

Joined Jan 12, 2018
57
Thanks for the advice, yes I will try it...

BTW the AliExpresss one which has max 4 mins (max time, adjustable via switch / resistors) will take 30-50 days to arrive (Whaaa..?) so I will play with this one anyway for now.

These lamps are for show and for lighting the porch not for security... Cameras and signs and alarms are for that :) It is all about deterring, if they are determined and professional nothing will stop them, not even the owner with a gun (which in AU is a bit hard - maybe with a cricket bat lol - we can't even have Medieval crossbows ehehe)
 

Thread Starter

advarp

Joined Jan 12, 2018
57
Well, I have put a 330uF 25V capacitor in parallel with both existing caps in turn and... no effect! The delay time did not change at all. I did try several times in case the 330uF cap was 'stale' (not used for a few years) - but still same effect.

I did not try other caps but I wonder what would be the point... Was 330uF too large perhaps? I only have larger capacity ones and only one 16uF or so. I thought the effect would be more visible with the 330uF plus the existing cap. But nothing./..

I will try to identify some resistor I could change but they're tiny.

I have no idea why this did not work...
 

ebp

Joined Feb 8, 2018
2,332
Well crumbs!
Are there any parts that aren't visible in the photos you've posted? On the board with the IR sensor? Back of the main board? Any little mounds of black epoxy right on the surface of the board?
 

Thread Starter

advarp

Joined Jan 12, 2018
57
haha - looking for a PIC? No, no blobs of epoxy. Well there are a lot of resistors on the small PCB that holds the PIR sensor itself. Will have a better look, maybe there.. don;t think so though.

BTW from the photos, the AE model has a series of resistors which are switched through a mechanical multi-way sliding switch and they are all located on the main board . The adjustable model has double sided PCB plating while this one is one-sided. I might be able to reconstruct the schematic. As far as I can see there are no PICs or anything else - how do they achieve the timing, just RC? Could be.

EDIT: this PCB is also double sided
 
Last edited:

ebp

Joined Feb 8, 2018
2,332
Actually I thought there probably wasn't anything or you would have posted photos already.
The is a widely used integrated circuit for PIR detectors. They use multiple capacitors for setting time delays. I'm amazed there is no evidence of a special function IC on the board. Perhaps the function is integrated right into the sensor package.
 
Top