Micro switches are to sense position, control the SSR or contactor through those.
I wonder how many reputed companies use really cheap stuff, at least here the parts seem genuine.
I wonder how many reputed companies use really cheap stuff, at least here the parts seem genuine.
This is not a temperature setting device. It's sole purpose is to open the flue damper, send a signal to the burner that the damper is open, and then, five minutes after the burner shuts down, to close the damper. The purpose is to prevent winter cold from entering the boiler from the chimney. It's been saving me several hundred dollars a year on fuel oil. plus it keeps the shop warm by stopping the drafts from coming down both flue pipes.
It is poorly implemented, for sure. But I don't have the time, space, nor the will to completely design something new for this. It's a tight package and the requirement to fit the 3 microswitches against the cam of the motor are pretty rigid. I'd rather replace the failed parts and add an MOV.
It is poorly implemented, for sure. But I don't have the time, space, nor the will to completely design something new for this. It's a tight package and the requirement to fit the 3 microswitches against the cam of the motor are pretty rigid. I'd rather replace the failed parts and add an MOV.
The mystery for me is how the low voltage semis are powered from line voltage - some way cheap and dirty, to be sure. I suspect there may be a fusible resistor instead of a fuse, if there's even that level of protection.
To my knowledge, the answer at post #2 is correct. 2904 is the St Microelectronics marking for LM2904D , LM2904DT and LM2904PT. This is listed near the end of the datasheet, section 8 "ordering information". The alphanumeric code next to the ST logo is the trace code: manufacturing plant, year and week, and process type. I don't have a cheat sheet to decode it.
LM2904 is plausible on this application. A quick check of the circuit traces after removing the part should confirm or disprove. Also check the resistors and capacitors connected to the IC pins, if possible. Values may have drifted.
My experience suggests that the power device (triac on this case, it seems) is failing first, and the circuit driving it is the next victim. If the circuit supply is a capacitor dropper, this is the weakest link and after 5-10 years it is degraded anyway. On your circuit, the two 4300 ohm power resistors in series followed by a diode may be the dropper. This may work if the line supply is about 120V and the current draw is about 10-12 mA. Check if the diode has become open.
LM2904 is plausible on this application. A quick check of the circuit traces after removing the part should confirm or disprove. Also check the resistors and capacitors connected to the IC pins, if possible. Values may have drifted.
My experience suggests that the power device (triac on this case, it seems) is failing first, and the circuit driving it is the next victim. If the circuit supply is a capacitor dropper, this is the weakest link and after 5-10 years it is degraded anyway. On your circuit, the two 4300 ohm power resistors in series followed by a diode may be the dropper. This may work if the line supply is about 120V and the current draw is about 10-12 mA. Check if the diode has become open.
The dropper resistors and the diode are the current limiting devices indeed. This is common on domestic appliances. The purpose is to confine the fault to the board, avoiding further damage to other parts. A repair is done by throwing away the whole board or subsystem. For the manufacturer, component level repair is not cost effective on home appliances; boards are discarded even if the fault is only a bad solder joint. There is often a proper fuse somewere else in the device to comply with safety standards.
I appreciate reading the possible fixes for such a problem. Then I wonder while glancing at a book about how some of these things were done long ago, if we haven't taken too must technology a bit too far past its real value.
Sorry to butt in here, it is that I see too much of our 'modern' design and acceptance of these advances. The function behind the circuit is a very good one. Maybe it can be done another way.
Sorry to butt in here, it is that I see too much of our 'modern' design and acceptance of these advances. The function behind the circuit is a very good one. Maybe it can be done another way.
In this case, the "fuse" is the PCB trace. 🙂 Thanks for the educated guesses on how this thing works. Sounds about right to me.
I have these two items in my cart at Mouser. The device "D2" next to "Q1" is a three terminal device. I cannot make out any readable print on it Hopefully the problem is Q1 or maybe D1.
Reading D2 gives conflicting readings depending what I use. It reads about 2 megohms on an ohmmeter. Component test shows a short between the left two pins. Diode checker shows the same. I assume the pin on the lower right of D2 is not connected to anything. I'm not all that familiar with SMD parts (I'm a dinosaur from the vacuum tube era), so maybe contemporary folks here might be able to ID this part better than I can.
I have these two items in my cart at Mouser. The device "D2" next to "Q1" is a three terminal device. I cannot make out any readable print on it Hopefully the problem is Q1 or maybe D1.
Reading D2 gives conflicting readings depending what I use. It reads about 2 megohms on an ohmmeter. Component test shows a short between the left two pins. Diode checker shows the same. I assume the pin on the lower right of D2 is not connected to anything. I'm not all that familiar with SMD parts (I'm a dinosaur from the vacuum tube era), so maybe contemporary folks here might be able to ID this part better than I can.
When I'm repairing boards like this one, and reassembling the device and testing it does not take too long, I often go for the lazy route. First a quick check of supply components just to be sure that there are no obvious shorts or opens, then replace the most likely culprits that are cheap and easy to source and desolder. On this case, the two parts you are already about to order, and the supply diode because sometimes they test good at component tester but are faulty when current should go trough them. It is 99,9% of times a 600V or 1000V 1A standard rectifier. If the fault is not solved, then actual troubleshooting is needed. At this point the previously replaced parts may be dead again, so order two of them.
The issue would require a thread by itself. But I reassure you that on a few price-insensitive niche markets that are not consumer electronics, things are still designed and built to be durable and repairable. The board discussed on this thread is actually a good one, from a repairability and dependability point of view. It is lacking surge suppression and conformal coating and it is built on the flimsy side but there are no custom parts and no attempts to make repair difficult on purpose. I bet that it has been designed in USA, probably a long time ago or at least from someone that still remembers how to make circuits without microcontrollers. A design made today would probably have: a power supply done with a capacitor dropper, because the control logic would consume 10 times less current and then a capacitor is cheaper; a 8 pin 20 cents microcontroller glued to the board with a drop of black goo; the triac and/or relays; and a few passive components. Quick and easy to design and test, even by comparatively inexperienced people, cheaper to manifacture due to the simplified bill of materials, less reliable but still good enough for today's end customer expectations, and almost impossible to repair when the microcontroller will die. By the way it is a when, not a if. The microcontroller program is stored in a flash memory inside the chip and will erase itself after several years, or by some unlucky supply surges happening just at the right time while the program is cycling.
I can only replace what I can identify in this case. The big diode, D1 does produce a good curve, so I suspect it is okay. The triac and the op amp are my leading suspects. I'll try replacing those first,
I have two units that are both dead, so I may be repeating the process if I can get this one working.
I have two units that are both dead, so I may be repeating the process if I can get this one working.
If you intend to put a MOV on the board for increased protection, you will need to locate it after a fuse or other suitable protection, as MOVs tend to fail short. Maybe it's time to put an in-line fuse holder in series with the hot lead feeding the board.
pcan,
Thank you for a comprehensive and non judgmental response. I think that this becomes the true nature of communication in its better sense. I can see your points clearly enough and also respect them to the degree that I understand the concept. My wavering with so many things progressive was started during my career with the phone company a lifetime ago. Most everything was built to last for decades. It was of course, in spite of the cost of such thinking. As time went on change occurred and I witnessed bit by bit sub quality creep in. The rest you can assume for yourself and still get the point.
Back to our scheduled programming.
I could put one of those 1 amp micro fuses somewhere on the PCB by cutting a trace and rewiring a bit.
Or you can use a wall wart / external supply, and feed DC to board at the right place.
That gives room to add protection.
Just a suggestion...
That gives room to add protection.
Just a suggestion...
I'd rather not add complexity (and resulting failure points) to the system.
I've ordered the triac and the op amp just now. Arrives on the 27th. Will report back results then.
I've ordered the triac and the op amp just now. Arrives on the 27th. Will report back results then.
That triac is a 1 amp thing in TO-92, used here for ceiling fan speed regulators, usually fired by a diac.
Look at the motor load, service / grease it as well.
Check the diac, if it fails it will not trigger the triac.
And a wall wart is a lot better than a capacitor dropper supply.
Look at the motor load, service / grease it as well.
Check the diac, if it fails it will not trigger the triac.
And a wall wart is a lot better than a capacitor dropper supply.
After replacing the triac and op amp, I bench tested the unit and it appears to work.
It has a five minute delay to close the damper. When AC is applied to the Sense wire, it opens the damper. I've reinstalled it.
Now I have a newer replacement unit which failed within 2 years of replacing this one.
As can be seen, the circuit uses a 16 pin IC this revision. Part numbers have been sanitized. Any guesses on what that 16 pin IC is?
If I can fix this one, I'll have a backup spare.
It has a five minute delay to close the damper. When AC is applied to the Sense wire, it opens the damper. I've reinstalled it.
Now I have a newer replacement unit which failed within 2 years of replacing this one.
As can be seen, the circuit uses a 16 pin IC this revision. Part numbers have been sanitized. Any guesses on what that 16 pin IC is?
If I can fix this one, I'll have a backup spare.
Unfortunately, that 16-pin IC may be a microcontroller with proprietary code burned into its ROM. I'd try shotgunning the triac anyway, and adding an MOV for surge protection (with a fuse in front of it).