Measure it and check but if that is orange, orange, black, silver and brown it is 3.3 ohm which would place a near short on the 34 volt line.
It could all be bad news having used something so low because that will have put a high voltage (the 34 volts) onto the 19 volt rail and that could easily have damaged the LC1818 chip 🙁 R97 is all that stood in the way which is why it went up in smoke.
It could all be bad news having used something so low because that will have put a high voltage (the 34 volts) onto the 19 volt rail and that could easily have damaged the LC1818 chip 🙁 R97 is all that stood in the way which is why it went up in smoke.
It should be a 20 volt part. You loose around 0.6 volts across the base emitter junction of Q93. Both those small transistors could be zapped.
When you replaced the capacitors, you may have put one in with reverse polarity. That could have been the first "new" fault. Maybe check this before further repair.
The 3.3 Ohm short should have fried the LC7818 IC, as it is +-20 Volt rated. The advice "use some resistor" was made for someone with at least a bit more practice in electronics repair. A 1k Ohm should have worked, even as this was not neccesary, looking at the measured voltages.
With remote repair help it is always diffilcult to sense the guy on the other side, his tools, skills etc.
People that only do as said are the easy task, those realizing own ideas without prior asking are problematic.
Here is a quite reliabel source for this old IC, they have a data sheet too:
https://www.reichelt.de/spezial-sch...qDRXitemzAohtkFIoueAIRo9_rmBL3MgaAnKlEALw_wcB
The 3.3 Ohm short should have fried the LC7818 IC, as it is +-20 Volt rated. The advice "use some resistor" was made for someone with at least a bit more practice in electronics repair. A 1k Ohm should have worked, even as this was not neccesary, looking at the measured voltages.
With remote repair help it is always diffilcult to sense the guy on the other side, his tools, skills etc.
People that only do as said are the easy task, those realizing own ideas without prior asking are problematic.
Here is a quite reliabel source for this old IC, they have a data sheet too:
https://www.reichelt.de/spezial-sch...qDRXitemzAohtkFIoueAIRo9_rmBL3MgaAnKlEALw_wcB
Good information, a little late i´m afraid. The resitor i used was around 4 Ohms. (Lack of knowledge on my part).
I will order some of these, if i can´t find any 2603´s.
Ok, it was some green goo on one of the legs on C773, otherwise the PCB-4 board was clean.
About 4 Ohms, i´m afraid.
Is there a way to check if the LC7818 i ok?
I guess i will have to replace the R97 and Q93 and Q97 before any more checking?
How do i check the D91 where it sits on the board?
The Zener isn't very critical as such. Any low power 400mw to 800mw 20 volt type is fine. As you can see from your spec sheet, the tolerances are quite wide.
The BZX79 series used to be common.
https://uk.rs-online.com/web/p/zene...p.ds&msclkid=9ac15b4ec8cf14812e1c3c6a68250f70
If you use something like the BC546 remember that the pinout of the European BC's is different to the Japanese 2S series.
The only way to check the IC is to see if it works and does not draw excess current and we suspect that before this happened that something is drawing to much current.
Remember the diagram is wrong and the diode and cap are shown the wrong way around. The negative end of the cap goes to ground and the end of the diode with a stripe goes to the FET.
Drawn incorrectly:
The BZX79 series used to be common.
https://uk.rs-online.com/web/p/zene...p.ds&msclkid=9ac15b4ec8cf14812e1c3c6a68250f70
If you use something like the BC546 remember that the pinout of the European BC's is different to the Japanese 2S series.
The only way to check the IC is to see if it works and does not draw excess current and we suspect that before this happened that something is drawing to much current.
If you remove R97 and refit the FET you should see 20 volts across the diode. You can also fit the 2700 ohm resistor in place of the FET if you want to try it. If the Zener reads short circuit it is zapped, if it doesn't read short it may or may not be OK but I would replace anyway seeing as it has been subject to a major overload.
Remember the diagram is wrong and the diode and cap are shown the wrong way around. The negative end of the cap goes to ground and the end of the diode with a stripe goes to the FET.
Drawn incorrectly:
I´m sure the polarity is correct on these capacitors, i put the new ones in the same way to old ones sat, but i will check again anyway.
I'm guilty of lack of practice on these things, but i´m learning. Is there a way to check if the LC7818 IC is actually fried?
Thanks for the link. Do they have all the other parts i need to complete the circuit too?
It helps with info what parts to use when troubleshooting and info for equivalent parts. Replacing parts is not a problem here 🙂
Don't struggle trying to get exact same parts as fitted. For all the transistors you can use common parts in daily use.
For Q91, the 2SB1015 you can use a TIP42C. Same pin out.
Use BC546 for Q93 and Q97 but watch the pinouts. The base is the middle lead on these.
For Q91, the 2SB1015 you can use a TIP42C. Same pin out.
Use BC546 for Q93 and Q97 but watch the pinouts. The base is the middle lead on these.
Looks like the diode and cap are sitting correctly on the board.
I have placed an order for these parts, i hope they´ll work.
Is the onsemi J112 good for Q95?
.
Ok, i can solder the original back for testing and if it works i´ll let it sit put.
That is if it doesn't destroy anything IF it's bad?
Is there some way i can test the original Q95 with my DMM´s?
That is if it doesn't destroy anything IF it's bad?
Is there some way i can test the original Q95 with my DMM´s?
I took some Ohm measurments on the SK246 and compared with the new SK364 and both were open S to D.
Connecting G to either S or D on the 246 gave 300 Ohms and for the 364 34 Ohms.
I guess the 364 is a bad candidate for replacement?
And as i think you are right about the SK264, not being bad, i will put it back for now.
Connecting G to either S or D on the 246 gave 300 Ohms and for the 364 34 Ohms.
I guess the 364 is a bad candidate for replacement?
And as i think you are right about the SK264, not being bad, i will put it back for now.
JFET's are a bit unusual to test. On diode range on your meter an N channel JFET like these should read like a diode from gate to drain and from gate to source. Red lead to the gate if you are using a normal digital meter.
Drain to source can read from open to quite a low ohms value and this depends on stray charge the gate has picked up... JFET's are voltage driven unlike ordinary transistors (which are current driven) and the gate has near infinite resistance at DC and so picks up charge.
Also for most JFET's like these the drain and source leads are interchangeable as the device is truly symmetrical. That only applies to JFET's and not to any other type of FET.
Drain to source can read from open to quite a low ohms value and this depends on stray charge the gate has picked up... JFET's are voltage driven unlike ordinary transistors (which are current driven) and the gate has near infinite resistance at DC and so picks up charge.
Also for most JFET's like these the drain and source leads are interchangeable as the device is truly symmetrical. That only applies to JFET's and not to any other type of FET.
Ok, now the original Q95 is back in as it tested good.
Measured R101 (still on board) 6,4 Ohms.
Measured R95 (still on board) 330 Ohms.
Out is R97 (burnt),
Q97 tested 1,6v at E+C and 0 at E+B, (burnt),
Q93 tested 0,7v at E+B and 0,7v at C+B (good),
D91 tested 0,0001 i both directions (burnt).
I will pull the Q91 for testing too.
New parts will arrive next week.
Measured R101 (still on board) 6,4 Ohms.
Measured R95 (still on board) 330 Ohms.
Out is R97 (burnt),
Q97 tested 1,6v at E+C and 0 at E+B, (burnt),
Q93 tested 0,7v at E+B and 0,7v at C+B (good),
D91 tested 0,0001 i both directions (burnt).
I will pull the Q91 for testing too.
New parts will arrive next week.
This is a moderately critical part because the value sets the current limit. When the voltage across this resistor exceeds around 0.6 volts Q97 turns on and pulls the regulator drive current down.
Assuming 4.7 ohms we get a limit of 0.6/4.7 which is 127ma. With the resistor at a measured 6.4 ohm we get a limit of 94 milliamps. Is it enough? Your meter leads will have around 0.2 to 0.3 ohms resistance unless they are specialist high current leads (low resistance) so the true value is more like 6.1 ohms. Either way, it has suffered but you can still work with it for now.
These are basic go/no go tests but are far from definitive as they do not show leakage of impaired gain etc.
Well and truly zapped. On a positive note it failing short means it clamps the voltage but on a negative note Q97 being zapped means a high current has flowed into the 19 volt rail and that I'm afraid points back to the LC chips. being possibly zapped.
All you can do is rebuild the regulator and I would recommend you use a DBT (bulb tester) initially to limit current. The bulb would need to be a very low wattage perhaps like a fridge bulb. Check the temperature of Q91 immediately after switching on and if it starts frying switch off immediately.
I pulled out the Q91 and found 0,6 v B to E and 0,6 v C to E and 0 B to C With my Fluke 11.
Seems good but i will replace with TIP42C anyway.
Will this work for DBT? (If i can find a proper bulb)
Seems good but i will replace with TIP42C anyway.
Will this work for DBT? (If i can find a proper bulb)