Has Anyone Successfully Repaired Sipea 0218 Delay Relay?

[aarpcard]

I tore into the electrics again this past weekend. One problem I discovered was the Sipea 0218 Seat Belt delay relay was stuck open.

I took it apart and discovered that the heating element had burned out. The element itself is a tiny strand of filament wire insulated with some kind of fiberous weave, which is wrapped around the armature of the relay. It looks to me like the fiberous weave broke down and caused the filament to short on the armature, overloading it, and blowing the filament (and possibly warping the armature since the failure mode for this relay should be stuck closed.)

IMO this relay is a really poor design, as the heating filament is on whenever the car is on - so it's going to wear out repeatedly. I couldn't find a New relay or new equivalent relay. Best I could find was a used (but working) relay on MWB.

I'm not a fan of buying a used Sipea 0218, because it's obviously just going to blow again at some point in the nearer future than if I could get my hands on a new one.

I've toyed with the idea of trying to source some filament wire to repair the relay, but that will probably be a difficult endevour as it will need to be insulated, and have roughly the same properties as the original (or else the relay will take way to long to open, or open way too quickly, or never open) and I have basically nothing to go on.

Has anyone successfully repaired this relay or found an equivalent new replacement?

 

[lookforjoe]

Got pics?

I would take that relay & lob it as far as possible, and bypass or remove the circuit. Failing that, I may have one you can have for nothing, since I removed nonsense like that from my car - post a reference pic though.

What is the delay aspect of the relay, can't visualizw what that would accomplish regarding the seat belt warning system.

 

[Dr.Jeff]

I'm not familiar with the functions of this particular relay, so my thoughts here may be off base. But perhaps if you know the relay's parameters you can utilize some Bosch item (or other quality component) in its place by adapting the connector plug with jumper wires. Also, is it possible to rewire the circuit with an additional relay to convert the function to normally cold instead of hot?

 

[aarpcard]

Got pics?

I would take that relay & lob it as far as possible, and bypass or remove the circuit. Failing that, I may have one you can have for nothing, since I removed nonsense like that from my car - post a reference pic though.

What is the delay aspect of the relay, can't visualizw what that would accomplish regarding the seat belt warning system.

Not my pics, but taken from google (nowhere near my car atm)

You can see the filament wrapped around the armature. The fibrous material is presumably a heat resistant insulator. On mine, I can see that the fibrous material wore away and shorted the filament on the armature, blowing it. It's hard to tell in these pics, but the filament is extremely thin wire (I doubt I could actually photograph it and get it to be viewable).

Below is taken from the wiring diagnostics manual. The Relay in question is circled in red:

Basically it controls the proper operation of the warning chime (whether it be for the doors, or ignition key, or seatbelt unbuckled). It also controls the operation of the parking brake light and seatbelt light on the dashboard.

If it fails open, then the seat belt light on the dash will not work, and the chime will not function when the seatbelt is unbuckled.

If it fails closed, then the chime will work when the seat belt is unbuckled, but the door jam switches will no longer trigger the chime. Additionally, the seatbelt light and the parking brake light on the dash will be stuck on.

The time delay function results in the relay being closed for 5 seconds after the key is turned to the run/start position. After 5 seconds, it opens. If your seat belt is unbuckled during those 5 seconds, then the chime will go off. The seat belt (and parking brake light) on the dash will be on for those 5 seconds, then they will turn off (unless the parking brake is on - then the parking brake light will stay on) regardless of the state of the seat belt buckle. During those 5 seconds, the door jam switches will not operate the chime. After the relay opens, they will.

The design of the circuit isn't great since in normal operation, unbuckling your seatbelt after the first 5 seconds will result in no warning light or chime. Also if your belt is buckled, but you have the doors open during the first 5 seconds, the chime will also not go off.

I'm torn between chucking the circuit and fixing it. The electrics in my car were/are in a nasty state. Literally almost every circuit was non-functional or intermittent. I've fixed the majority of the electrics - the goal is to fix/restore them all. Now most things work, instead of nothing working (and the wiring is safer and cleaner too). There were several other things wrong with this circuit that I've already addressed - this relay is the last failure point. As I said above, the circuit is a poor design from a functional standpoint, but part of me wants to get the car back to fully functional stock condition.

I'm not familiar with the functions of this particular relay, so my thoughts here may be off base. But perhaps if you know the relay's parameters you can utilize some Bosch item (or other quality component) in its place by adapting the connector plug with jumper wires. Also, is it possible to rewire the circuit with an additional relay to convert the function to normally cold instead of hot?

I'd like to use a modern quality replacement - even if it means using jumper wires, however an equivalent is proving hard to find (but it could also be I'm not looking in the right places.)

As for rewiring the circuit - I thought about that. Because the relay is triggered by a heater, some kind of latching functionality would have to be built in. Something like:

-Turn Car On -> 12V applied to filament -> 5 seconds later relay opens -> Relay opening also disconnects filament from 12V, and triggers a D-Latch circuit that holds the relay outputs open.

Without drawing it out, I think it could be done, but would require two additional normal relays and a latch. Maybe it would be better to just design and build a solid state circuit that mimics the Sipea 0218 - probably just need a 555 timer in single shot mode, a latch, and some beefier mosfets on the outputs . . .

 

[lookforjoe]

Sounds like you have the where-with-all to redesign the relay to modern standards, using solid state instead of bi-metallic heater contacts If you're trying to maintain/resurrect stock devices, I can understand keeping that in place. Just building one relay that mimics the function, even if it's slightly longer duration, would seem the most practical. The chime/warning in the current 500 is most horrible, it stays on for about 5mins, and gets more insistent the longer you ignore it.

 

[aarpcard]

Sounds like you have the where-with-all to redesign the relay to modern standards, using solid state instead of bi-metallic heater contacts If you're trying to maintain/resurrect stock devices, I can understand keeping that in place. Just building one relay that mimics the function, even if it's slightly longer duration, would seem the most practical. The chime/warning in the current 500 is most horrible, it stays on for about 5mins, and gets more insistent the longer you ignore it.

Yeah I think I'm going to look into building an equivalent circuit. I looked a bit harder and the only similar relay (in terms of operation) was one by Hella and cost ~$52 new . Plus the closest options for delay were <2.5 second or >10 seconds - which are substantially too short and long.

I agree most chimes on cars (especially more modern ones it seems) are annoying and incessant.

However I kind of actually like the chime on the X1/9. It's an electromechanical bell that isn't too loud and isn't too high pitched. IMHO it adds to the character of the car.

It's also nice to have an audible reminder that a door isn't fully latched or you left your key in the ignition (something which I have a bad habit of doing lol).

 

[Dan Sarandrea (Phila)]

Looking at the pics shown when searching in Google, the innards of a Ford seat belt gizmo look like they are made the same way as the Sipea. Maybe a jumping off point? Also, FIAT parts sources show a differently connected "seat belt delayer" for x1/9 and 124 from '74 to '78, it has a pigtail with a connector housing on the end, maybe the innards and circuit design are close enough to work with?

 

[Dr.Jeff]

Bosch has a relay with a variable/settable delay, but it's around $60-70. And you would still need to make all the related functions work with it. I don't think rebuilding the old one is likely to work properly, but that is just an assumption. Sounds like the custom PCB idea is best. Or just ditch the whole thing.

 

[c snapp]

I took my relay out when I first acquired my car.
I don't need the government to remind me to fasten my seat belt.

 

[aarpcard]

So I built a solid state version of the relay today. I was able to simplify the circuit enough to fit inside the original relay casing.

Below is the schematics for the circuit. It's essentially a monostable 555 timer that is triggered on power on. "+", "-", "C", and "C1" correspond to the pins on the original Sipea 0218 relay. The output of the 555 timer operates two power N-channel Mosfets. You can use any kind of NPN Mosfet transistor here, but the ID rating has to be at least 600mA and the VDS rating must be at least 20V.

I set the delay to ~8 seconds, but it can be changed to whatever you want by changing the values of R1 and C1. Just put the relay in the car and it's working perfectly (and it should have a much, much longer life span).

I don't have great pictures of the final relay, but here is what I do have:

It almost completely fits inside the housing. I used tape to insulate the casing from the PCB. I also used tape to attach the relay cover to the base (no picture) - doesn't look great, but it works. If you wanted, you could 3d print a better housing - maybe I'll do that later.

Pretty much as dense as you can get with DIP and hand soldering. . . . I potted the bottom of the board in hot glue after I was done.

 

[Dr.Jeff]

I think he said something about a flux capacitor.

Very impressive, really nice job. I wish I could solder like that. Amazing that it is small enough to fit inside the relay case. Thumbs up!

 

[bbrown]

Nice work. Nice design too. + 1

 

[lookforjoe]

Nice work!

I’m curious- where do you get spec sheets that indicate what resistor and capacitor values to use to set timer value? I ask because I used a Volvo 6 min heated mirror timer relay for my engine bay fan, but had to add a secondary relay to govern the momentary contact switch originally used to trigger the relay circuit. Probably could make my own relay that doesn’t need external switching....

 

[Dan Sarandrea (Phila)]

So I built a solid state version of the relay today. I was able to simplify the circuit enough to fit inside the original relay casing.

Below is the schematics for the circuit. It's essentially a monostable 555 timer that is triggered on power on. "+", "-", "C", and "C1" correspond to the pins on the original Sipea 0218 relay. The output of the 555 timer operates two power N-channel Mosfets. You can use any kind of NPN Mosfet transistor here, but the ID rating has to be at least 600mA and the VDS rating must be at least 20V.

I set the delay to ~8 seconds, but it can be changed to whatever you want by changing the values of R1 and C1. Just put the relay in the car and it's working perfectly (and it should have a much, much longer life span).
View attachment 5954

I don't have great pictures of the final relay, but here is what I do have:

It almost completely fits inside the housing. I used tape to insulate the casing from the PCB. I also used tape to attach the relay cover to the base (no picture) - doesn't look great, but it works. If you wanted, you could 3d print a better housing - maybe I'll do that later.
View attachment 5955


Pretty much as dense as you can get with DIP and hand soldering. . . . I potted the bottom of the board in hot glue after I was done.
View attachment 5956

Hi Tyler,

When it comes to electrical and electronics, if Bob Brown sez you done good, then Yep, you done good!

 

[aarpcard]

Nice work!

I’m curious- where do you get spec sheets that indicate what resistor and capacitor values to use to set timer value? I ask because I used a Volvo 6 min heated mirror timer relay for my engine bay fan, but had to add a secondary relay to govern the momentary contact switch originally used to trigger the relay circuit. Probably could make my own relay that doesn’t need external switching....

Thanks!

See attached data sheet for more info. Page 9 deals specifically with monostable operation. According to the datasheet, the pulse width (on state delay) is governed by T(in seconds) = 1.1 * Ra (in Ohms) * C (in farads). For the ~8 second delay, I used 22uF and 330kOhms. If you want a 6 minute delay, you'll need fairly high resistance or capacitance values. I'd try to keep the capacitance reasonably small (maybe around 100uF - 150uF) as I've had some issues with 555 timers operating correctly in the past if the timing capacitor is too large. You could also use a potentiometer in place of the resistor to change the delay at your whim. The circuit I came up with, triggers on power up - the filter comprised of R2 and C3 on the trigger hold the trigger low on power up and then keeps if high indefinitely after C3 charges. If you want it to act more like a normal relay, you'd remove that circuitry.

Hi Tyler,

When it comes to electrical and electronics, if Bob Brown sez you done good, then Yep, you done good!

Haha I'm honored - I figured it must be a decent design if it got his seal of approval!