 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Electronics : capacitor availability.
Page 1 of 2   |
|||||
| Author | Message | ||||
Dinosaur Guru  Joined: 12/08/2011 Location: AustraliaPosts: 311 |
Hi All I am repairing a AMI Rowe R-82 Jukebox for a friend. There are 3 capacitors of which 2 are suspect, one definitely blown. Replacing all three. All are axial electrolytic. They are all part of the main power supply for which there is no circuit in their repair manual. The PS delivers: 28vdc 3A fuse -27vdc 1A fuse -7vdc 2.5A fuse The first: Sprague 70023310 1250UF -50vdc Physically 50mm long & 25mm dia. Sticker says Aug 1977. Should be easily replaced. The second: Mallory 3000MFD 15vdc S-235K 235-7731B Physically 40mm long 22 mm dia. This one is a puzzle to me. 3000 milli Farad ? (3000,000 micro farad) in such a small package ? What is the chance of the marking meaning 3000 MicroFarad ? The third: Standard 220uF 80vdc Does anyone here have any idea on the second cap ? Regards Regards Hervey Bay Qld. |
||||
| pd-- Senior Member  Joined: 11/12/2020 Location: AustraliaPosts: 122 |
Like this one https://www.ebay.com/p/2085807279 |
||||
phil99 Guru Joined: 11/02/2018 Location: AustraliaPosts: 2135 |
In the days of valves MF was uF and MMF (micro-micro-farad) was pF. They had no need for milli-farad caps, and didn't bother with nano-farads. 3300uF 25V will give you a safety margin. As modern caps. are smaller there should be room. Edited 2023-01-08 16:48 by phil99 |
||||
| Dinosaur Guru Joined: 12/08/2011 Location: AustraliaPosts: 311 |
Hi All Many thanks for the replies. @pd-- it is actually a round barrel type with central lead at each end. @phil99 I actually found the circuit and your suggestion is what I concluded as well. The circuit is a full bridge with the cap across it's output.Then they use a discreet component voltage regulator circuit using 3 transistors resistors and another single 0.1mfd. Will trip to our local Jaycar and see what I can get. REgards Regards Hervey Bay Qld. |
||||
| Dinosaur Guru Joined: 12/08/2011 Location: AustraliaPosts: 311 |
Hi All First attempt at attaching a .pdf Having difficulty finding a replacement transistor which are listed in the spares list as "Transistor,Silicon,NPN (Motorola SPS6979; Spraque TZ8630; G.E.X32B 4682" The part does not come up in the Motorola cross reference list. On the circuit they are Q5 & Q6 which also have the number S 300B with them. Would anyone hazard to guess an alternative looking at the circuit. The Voltage rails are +30 and -30vdc . The center rail (base of Q7 pair) is suppose to be driven to about 0v with no input on terminal 3 on left of drawing. Would love someone to tell how and what Q5 & Q6 are suppose to do. Regards Screenshot.pdf Edited 2023-01-15 15:45 by Dinosaur Regards Hervey Bay Qld. |
||||
| pd-- Senior Member Joined: 11/12/2020 Location: AustraliaPosts: 122 |
a back to back zener diode. it would limit the voltage that the base of Q7 is exposed to from pin 3 & pin 4 you would probably need to go back to scanned copies of old parts catalogs to find them. its probably easier to experiment & start with sum 6v zeners |
||||
| phil99 Guru Joined: 11/02/2018 Location: AustraliaPosts: 2135 |
Yes, the purpose of Q5 & Q6 is to limit the audio amplitude. They probably used B-E reverse breakdown as it clips the audio more gradually than zenners would, producing fewer harsh harmonics. Most small signal transistors should be ok there, eg BC547 or 8 or 9. Q7 is a thermally coupled matched pair. Haven't seen a through-hole one of those in a very long time. Surface mount dual transistors might be easier to find, though soldering legs on it would be fiddly. As a last resort two TO92 transistors from the same batch could be glued back to back. As the leads won't align with your PCB use strip-board to make an adapter. Any low-noise audio transistors with a Vce raring of 60V or more should work. Edited 2023-01-16 08:05 by phil99 |
||||
| Dinosaur Guru Joined: 12/08/2011 Location: AustraliaPosts: 311 |
Hi All Many thanks for the replies. I will start by replacing the Q5 & Q6 with the suggested transistors and see if it solves the problem. If so, then I will leave Q7 alone, otherwise the back to back option. Once again thanks for the responses. Regards Hervey Bay Qld. |
||||
| phil99 Guru Joined: 11/02/2018 Location: AustraliaPosts: 2135 |
Q5 and Q6 play no part in amplification. The only way they affect the signal is to limit the signal peaks so unless they are short circuit they won't be the problem. With the power off simply measure the resistance between the emitters. Repeat with the probes reversed. If you get a low reading either they or C15 might be a problem. Disconnect one emitter and check C15. If the resistance has risen C15 is probably ok. Leave Q5 or Q6 disconnected power on, apply a signal and see if it works. If not is the signal getting through R28 and C17 to the base of Q7? Your PDF indicates both base-emitter voltages of Q7 should be 0.55V The collectors should be about 30V above ground. The anode of CR9 should be -10V. If all those are correct Q7 is probably ok. Q3 and Q4 also play no part in amplification. They monitor the emitter currents of Q1 and Q2, applying extra negative feedback to the righthand base of Q7 if the currents get too high. |
||||
| Dinosaur Guru Joined: 12/08/2011 Location: AustraliaPosts: 311 |
Hi All The Jukebox has two of these boards (Left & Right channel) and with one of them working I have been comparing signal levels. The top of the diagram is +30 and the bottom is -30 The only description of the board are these 4 lines: Q4 is part of the positive clamp circuit. Output device Q1 draws current through Resistor R2. (Should that be R1 ??) Q4 drops the base of output device Q1 to below R2, limiting current to a safe value. Q3 acts on the negative signal component in the same manner as positive clamp Q4. If I leave the Fuse F1 in (-31) the Q1 will overheat if I let it. So with the F1 open I have put a 1khz signal into both boards at 1v p2p. That is what was noted at left terminal 3. There is no current flowing through R25 and R20 is at 28vdc. However the 1V p2p drops to a dirty looking 0.3 v p2p at emitter of Q6 and the base of Q7 left side. I had not replaced C17 yet, but C15 is ceramic disc so probably ok. I was simply going to replace all transistors. I have already replaced all the electro caps and not touched other caps which are ceramic disc or mylar. However after finding the Q7 on line (2N2919) at $60 decided to go for the Q5 & Q6 at $0.30 from Jaycar and replace C17. Will follow your suggestion and test again in the morning. Edited 2023-01-16 16:30 by Dinosaur Regards Hervey Bay Qld. |
||||
| phil99 Guru Joined: 11/02/2018 Location: AustraliaPosts: 2135 |
After reading the description of Q3 & Q4 I see they act directly on Q1 & Q2 not via Q7. Q1 running at full current suggests Q2 could be short circuit. With power off measure its C-E resistance. With F1 removed the amp is unbalanced and most voltages will be wrong as R20 feeds the error back to Q7 and around the whole amp. Remove F2 as well. Power up and measure the voltage at R16, terminals 9 & 10. If the rest of the amp is working (input grounded) it should be close to 0V and voltages around Q7, 8 & 9 should be similar to the good amp. If Q1 is ok and the voltages around Q8 are wrong check it's C-E resistance, compare with other amp Q8. Edit. With no speaker connected Q1 can only get hot if Q2 is shorted as then the only current path for Q1 is through Q2. Edited 2023-01-17 08:07 by phil99 |
||||
| Dinosaur Guru Joined: 12/08/2011 Location: AustraliaPosts: 311 |
Hi All Q9 was short across C E so replaced and that's now ok. But The voltages are still way out. R20 (center rail) = 0.015V on good board and -10.16 on bad board. The Z1 diode pack is basically 3 diodes (4002) in series mounted on a small board which is then held against the heatsink with a spring loaded screw. I guess the 3 diodes create a higher voltage tolerance and provides additional resistance. The good board measures about 11 K ohm The bad one 10+ M ohm. Will get some and see how I go and let you know. Regards Hervey Bay Qld. |
||||
| phil99 Guru Joined: 11/02/2018 Location: AustraliaPosts: 2135 |
The role of the 3 diodes (Z1) is o provide a temperature compensated bias voltage between the bases of Q1 and Q2. As Q1 and Q2 heat up the voltage drops across the B-E junctions fall. By clamping the diodes to the heatsink the bias voltage they produce will fall a corresponding amount. A simple resistance check won't tell you if they are good or not. If your meter has a diode check function use that to test each of the diodes. The forward voltage should be 0.54V to 0.65V on most meters. |
||||
| Dinosaur Guru Joined: 12/08/2011 Location: AustraliaPosts: 311 |
Hi All After checking every Transistor, now none of them show a low resistance between E & C. Typically they are higher than 10k Ohm. So started taking voltages as per your suggestion. Some voltages that I am trying to get my head around. Rails are higher then marked +33v and -32.6v Bad board Good Board C17 + = -0.103v couldn's reach it CR9 = -9.6v -9.5v Q8 C = -10.3 +1.482 Q2 B = -10.16 +0.825 Q1 B = -10.78 -0.744 Q3 B = -9.56 +0.014 Q4 B = -10.03 +0.016 Q4 C = -2.54 +3.2 R13 = -21.8 -16 -10.8 -0.744 Top of C6 R19 = -10.29 +1.482 (same as Q8 C) R20 = -10.16 +0.31 T10 = -10.16 +0.015 Drop across Z1 = 0.3v 0.682v Without turning a live board over, some of the measurements are doubtful on which side of the Resistor it belongs. R19 a case in point when compared to Q8 Collector. EDIT: You were right again. Taking both boards onto the bench and using a current limited PS, i connected 5vdc to top of diode pack. Both boards drew 199mA and dropped the voltage to 2.5v. Edited 2023-01-17 15:58 by Dinosaur Regards Hervey Bay Qld. |
||||
| phil99 Guru Joined: 11/02/2018 Location: AustraliaPosts: 2135 |
Are the "Bad board" readings taken with both F1 & F2 removed and no speaker? With both Q1 & Q2 collectors disconnected it will be easier to see what the rest of the circuit is doing. Because R20 feeds the output voltage back to the input the effects of a fault go right around the loop. If the voltages change significantly with both fuses removed it suggests either Q1 or Q2 has a problem. Edited 2023-01-17 16:14 by phil99 |
||||
| Dinosaur Guru Joined: 12/08/2011 Location: AustraliaPosts: 311 |
Hi All Yes both board both fuses out. I will lift the Q! & Q2 out in the morning. After looking at my readings there are some missing which I will also do. Regards Hervey Bay Qld. |
||||
| phil99 Guru Joined: 11/02/2018 Location: AustraliaPosts: 2135 |
With both fuses removed the collectors are isolated so it should be possible to test them where they are. Check the resistance between B & E for Q! & Q2, they have internal resistors, value unknown but I expect something between 20 and 200 ohms. What ever it is, they should be similar. If that is the case the measurements you have taken indicate a problem elsewhere. To further test Q2 set your PS to 10V and 100mA. Connect neg. to terminals 9 & 10 and pos. to the output end of F2 fuse holder. There should be no current. Now connect a 1k resistor from pos. to base. The current should now be close to the limit you set and the voltage between base and emitter about 1.1V. For Q1 repeat with reversed polarity. . Edited 2023-01-17 17:03 by phil99 |
||||
| Dinosaur Guru Joined: 12/08/2011 Location: AustraliaPosts: 311 |
i All I started by taking the Q! & Q2 out on both boards and measuring all voltage points. Between the two boards there was an average difference of 50mV on various points so I think that is probably fine. Then replaced all the electro caps on the good board and tested again. Made very little difference. I started testing the Transistors last night but decided instead to order new ones and test while waiting for delivery. The Transistors are Q1 = MJE6041 & Q2 = MJE6044 (suppose to be 2N604* compatible) The best match and delivery was STM TIP102 x 5 for $10.90 on eBay (Sydney stock) STM TIP107 x 4 for $9.00 The only difference between them & original is B - E res = 5k + 150 ohm where the MJE6041 is 8k + 120 ohm. Don't think it will make any difference. The good board had a significant resistance difference between the NPN & PNP but are suppose to be identical according to the spec sheet. So, perhaps one of them may be damaged as well. Before the transistors arrive, I will quantify things and test with 1vpp signal again.Didn't want to nominate a faulty Q1 or Q2 until proper test was done. Also with both boards having the same transistors will make the final test more meaningful. Sofar so good. Regards Hervey Bay Qld. |
||||
TassyJim Guru Joined: 07/08/2011 Location: AustraliaPosts: 6098 |
I have been following this with interest and finally had to get involved. Can you measure the voltage drop across R24, R23 and R19. As it stands, it looks like Q8 is not turning on as much as it should. This is allowing the base voltages of Q1 and Q2 to be pulled too negative. As Phil99 pointed out, the feedback does make life difficult. Jim VK7JH MMedit  MMBasic Help |
||||
| Dinosaur Guru Joined: 12/08/2011 Location: AustraliaPosts: 311 |
Hi All TassyJim I swear I was going to find that when I started putting 1 vpp in again.  Good board Bad Board R24 = 0.629 0.651 (could be pot position) R23 = 0.491 0.001 R19 = 0.584 0.001 Note that C16 is 100pF ceramic disc. EDIT: Q8 is an MPSA56 Edited 2023-01-18 13:08 by Dinosaur Regards Hervey Bay Qld. |
||||
| Page 1 of 2 |
|||||
 Print this page |
