EEVblog Electronics Community Forum
General => General Technical Chat => Topic started by: SeanB on September 19, 2015, 04:31:15 pm
-
So, this SCR drive decided to stop working this week. Take the case apart, get the dust out ( not been opened for a long while, over 5 years IIRC) and get to the board. Apply power, no operation. Check the 2 fuses, they are fine. Either motor or board then. Look at board, and the power led ( one of 3 leds on the board) is not lit. So, power supply problem then, time to undo screws. Did a sanity check on the motor, brush length still fine, and it runs on 12VDC from a test lead acid pack.
Take board out, and do a physical check, nothing obviously blown or broken, so out with the multimeter, to check for broken traces ( board was made in the early 1990's, so I know this old board copper likes to peel or break) and nothing wrong. Check transformer, primary around 3k, secondary around 20R. Seems to be fine, time to trace out further.
Confusing in they use the diode tabs and the leads as jumpers, and clearance and creepage are close to nil, but not a worry as the entire board is at mains potential in any case. long ago I added reinforced insulation to the speed setting pot, not happy with there only being 1mm between the terminals and the case, though that is very well grounded to mains earth using a separate crimp lug, bolt and nylock nut. It got a 2mm plastic spacer added.
20VAC secondary, 1uF AC filter cap ( SCR makes a bit of line noise, so overkill it is) on the secondary is fine, and I start tracing out the circuit on the secondary side. 4 1N4004 diodes, bridge rectifier, correct. wire jumper, slightly misdrilled, but still soldered good enough, and a trace leading off to a resistor then to a quad opamp, so AC zero cross sensing for the SCR drive. Then another 1N4004, tests as a diode, and this leads via another trace off to the power LED.
After power LED trace leads to the main filter capacitor, 220uF 50V, and then onto the board and to a 78L08 regulator, obviously the voltage reference and regulated supply for the SCR drive.
Hang on, power for the entire board ( aside from the zero cross resistor) flows through a 5mm red LED? Is the LED working, so unsolder it after in circuit tests are inconclusive ( no light using DMM diode test) and try using a somewhat used CR16126 cell on hand. Still no light, so the LED is open circuit. Look around on the pile for another LED, and get the donor monitor board to give up a red LED, it already has donated the switches. Check Led, and it works using the Cr1616, so solder it in. Then screw the board back in, and gingerly apply power, the board dangling by the wiring in free air.
Power on, and turn up the wick, and the power led is doing a good impression of a spotlight, and the board is working. Might not last another 10 years at that forward current, so I need to do something about it. time to reduce LEDE current, so looked on the donor board, and spotted a 100R resistor. About right, will do, so placed in the only place available, under the LED where it fitted between convenient solder pads. Visible as the sole bodge on the board.
Connect up again, and this time screw the board back down before application of power, and run the motor at various speeds and with varying load ( hold shaft to drop speed and get speed control to give more power to maintain back EMF of the motor), and declare it working after a half hour of no smoke.
Look up the manufacturer on google, and on google maps see they are still around, though they have changed location and phone number since 1993/4, when this board was made. Phone up ,and actually get an answer, though he was not at work, but was at Swartkops Raceway watching a race.
Will call Monday to see if they supply this model or a similar cheap one, as both spare and for when the horrid Chinese speed control in another machine does what they regularly do and blow up. Those are easy though to fix, a new triac, repair the traces blown off the board and put in 4 1N5404 diodes to replace the 1N4002 diodes they originally came with, along with putting in 2 VDR units to catch the spikes. Seems to work, though the pots are hard to get, I normally use a more robust one, though I lose the redundant power switch on it.
-
Hang on, power for the entire board ( aside from the zero cross resistor) flows through a 5mm red LED?
Why was it designed like that? My rough guess is that designer decided to add power LED in late stage of design and run out of PCB space for parallel traces.
Or he/she was very confident of LEDs' quality - these days you can get poor quality LEDs for the price of good ones without even knowing about it (LEDs do not have any markings on them after all).
Also, the board looks pretty decent. Some older PCBs show damage at diode bridge area from extended heating. Diode bridge is inadequately cooled, so PCB itself and surrounding traces are turning black over time.
-
I will ask Monday when I call.
-
Dropping a little voltage to get under 30V for the 78L08?
Doesn't make sense with a 20v transformer but the one pictured appears to be 24v? Very curious anyway and am looking forward to hearing an explanation.
EDIT: Oh, transformer is not original.
-
Could be designed so it shows that current is being drawn. For example, if the LED goes out it indicates something else in the chain is faulty.
I had an audio amplifier that had the power LED in series with the MUTE pin on one of the main amplifier ICs. When the power switch was turned on, the MUTE pin is energised through the LED. It turned out, the MUTE pin had an internal current sink (~2mA IIRC) designed to discharge a capacitor attached to it for click/pop suppression - but these guys were using it to make an LED glow. From an unregulated 18V bus. Wonder how much power dissipation the IC's transistor was designed for.
-
Something that old I would hold suspect all Elko's on the board and replace them anyway. In an analog circuit temperature changes could throw them off value. Who knows the result. Cheap enough. My £1-50
-
There are only 4, and the actual value is not too critical on them, and ESR is not really too critical in any case so long as they are not open or leaky. Depending on the price of a new board I might replace the board and fix this as a spare, or replace them, but as they are still working I will not really worry much. I have the same brand of Siemens caps in 40 year old equipment, and the only reason they were changed is that the relay associated with the board burned out, and is no longer made. The old ones were still close to spec. There are some 40 year old ones there still in use as power supply capacitors, they are still fine, even though they have nasty power spikes applied by the relays and DC brake they drive. Of course being 470uF 200V units running on 50 VDC might help as well.