I hope this is the right sub forum (I doubted to post it in equipment and tools)
But I am reading a lot about troubleshooting the last few months and I found a few helpfull trics but I have some questions with it.
1. The light bulb test.
q: How to know if you've got the right bulb wattage
q:what if just 1 channel is shorted what will happen then
q; You can't test the amplifier while on a light bulb I think
2. Seperating the pre/ end amplifier
q: how can you seperate them if there's no bridge or a cable?
Other questions about symptoms:
1. Some people Always say: you need new capacitors, but what are really symptoms from bad capacitors: noise, cracky sound like (Ka-1500 problem - YouTube ) , bad bass
2. What are symptomps of bad transistors and in what stage?
Like if a output transistor is faulty, but a driver transistor is still good, whil you have good sound on headphones?
Last question: Do you know any good literature about the working of a amplifier? For example: I know a little bit of the driver stage/ the output stage(only what they are in terms), but but not enough so troubleshooting is a big problem for me. ( in an amplifier I am able to see what transistors are, but the transistors on the heatsink, i really don't know if they are all output or 2 are output and 2 are driver..)
But I am reading a lot about troubleshooting the last few months and I found a few helpfull trics but I have some questions with it.
1. The light bulb test.
q: How to know if you've got the right bulb wattage
q:what if just 1 channel is shorted what will happen then
q; You can't test the amplifier while on a light bulb I think
2. Seperating the pre/ end amplifier
q: how can you seperate them if there's no bridge or a cable?
Other questions about symptoms:
1. Some people Always say: you need new capacitors, but what are really symptoms from bad capacitors: noise, cracky sound like (Ka-1500 problem - YouTube ) , bad bass
2. What are symptomps of bad transistors and in what stage?
Like if a output transistor is faulty, but a driver transistor is still good, whil you have good sound on headphones?
Last question: Do you know any good literature about the working of a amplifier? For example: I know a little bit of the driver stage/ the output stage(only what they are in terms), but but not enough so troubleshooting is a big problem for me. ( in an amplifier I am able to see what transistors are, but the transistors on the heatsink, i really don't know if they are all output or 2 are output and 2 are driver..)
1. 100 watts is usually about right for most amps. An amp when idling (unless class A) draws little current so 60 or to 100watt is fine.
You can test with a bulb but only at very low volumes. Turn it up and the bulb lights drastically reducing the supply to the amp.
2. Separating pre and power if needed involves understanding the signal path and looking for a suitable component (such as a coupling cap) to remove. You need to understand the circuit though and remove something that won't upset the DC conditions.
3. Caps are a favourite item to replace but just because they are old doesn't automatically mean they are duff or past it. I have some 1970's Sony gear that is all original and just fine.
4. Transistors fail in many ways but a short circuit output device will draw excess current and blow fuses or blow other components round about until the amp fails. You wouldn't normally have sound on headphones if that were the case.
You can test with a bulb but only at very low volumes. Turn it up and the bulb lights drastically reducing the supply to the amp.
2. Separating pre and power if needed involves understanding the signal path and looking for a suitable component (such as a coupling cap) to remove. You need to understand the circuit though and remove something that won't upset the DC conditions.
3. Caps are a favourite item to replace but just because they are old doesn't automatically mean they are duff or past it. I have some 1970's Sony gear that is all original and just fine.
4. Transistors fail in many ways but a short circuit output device will draw excess current and blow fuses or blow other components round about until the amp fails. You wouldn't normally have sound on headphones if that were the case.
But i think you can't use the light bulb test with amplifiers that have little screens and such in it right?
Cause if you put things in an series arangement it will divide the power right?
The reason I am asking this is:
Mooly said to me that I needed to measure the dc drop across the B and E pins from my output transistors.
But if i measured it with an light bulb attached, will it give me the same results?
Like the 0.7v that I measured without light bulb will stay 0.7v with light bulb test?
Or is the light bulb test only for short circuits?
And if your amplifier has an speaker relay, do you need to short it to do the test?
Cause if you put things in an series arangement it will divide the power right?
The reason I am asking this is:
Mooly said to me that I needed to measure the dc drop across the B and E pins from my output transistors.
But if i measured it with an light bulb attached, will it give me the same results?
Like the 0.7v that I measured without light bulb will stay 0.7v with light bulb test?
Or is the light bulb test only for short circuits?
And if your amplifier has an speaker relay, do you need to short it to do the test?
Last edited:
The light bulb as a series AC limiter is only useful if the auxillary circuits of the amp make it to full voltage. That is the +-15 for any op amps, or the 24 for the relay driver, or if you have a vacuum florescent display, the voltage to that should be nominal. My 10 output transistor amp couldn't, so I ended up using a 1500 W room heater element in series with the AC. The purpose of the lightbulb or room heater is to limit the size of the explosions when stressed parts blow up. Without it I had dies blowing through the steel case of transistors that passed the diode test, and onto the ceiling.
The reason those of us with experience change >15 year old electrolytic capacitors is that they are wet systems sealed with rubber. Historic rubber is dirt. Just because the device wasn't used, doesn't mean oxygen wasn't attacking the rubber seal on electrolytic capacitors every day since production. how good are the tires on a car made the year of you amp? The air pressure tests okay. (like a good 2v capacitance check). Want to take those tires out at full speed on the motorway/freeway/autobahn? Nah, that would be stupid. Same with old electrolytic capacitors, they can be fine today and slime all over the PWB tomorrow. It is better for amateur solderers to get the amp working, then change the capacitors for reliability one at a time, to focus your eyes on exactly the mistake you just made soldering it in.
Output transistors are the root of a lot of failure in small band amps due to operating into shorted speaker wires, especially the 1/4" phone plug pulled partway out. Amps in organs or fixed wiring situations like churches or theaters, don't nearly have as many shorted output transistors.
The speaker relay contacts have to be jumpered, and the DC protection circuitry (if any) has to be defeated to determine if you have DC output on the speaker terminals. the right answer is less than 50 mvdc. Than put the protection back before final watt testing the amp and closing it up.
Diode tests on DVM's check the junction on a transistor at low voltage and low current. Forwards voltage drop on a good junction can be from .45v to .73v. (multiply by 1000 on my dvm, I use 2000 ohms scale). Backwards drop is 1999 or dash. I find the diode drop test of a DVM will determine that a transistor or diode is bad, but that they can still blow up after they passed that test. Mooly etc likes to measure gain, I don't have equipment that sophisticated. I found a Vce leakage test correlates to old transistors that don't blow up. It requires a voltage source, I use a car battery charger with a capacitor on it to make 17 VDC. I short the B to E on the transistor, then put a milliammeter between the collector of the transistor under test and the battery charger. + on collector for NPN, - on collector for PNP. New transistors don't leak very much current, old used ones do sometimes. when I eliminated the leaky transistors they stopped blowing up in the amp after 30 seconds to 2 minutes. You could do this test at 150 V or 250 V or whatever the Vceo of the transistor is, but voltage over 24 is dangerous to you. The 17 v test seemed to screen out stressed used transistors.
I load the speaker terminals of the amp with 10 ohm 450 W resistors, instead of speakers, for test. I put a cheap car radio speaker protected from DC by a non-polar electrolytic capacitor, across 5 ohms of the output resistor. When intermittant problems occur in the amp, while it is playing music, the speaker sounds weird, without actually ripping out the voice coil from the DC out at the same time.
Don't test areas of the amp over 24 V using two hands, current across your heart can stop it. Use a clip lead on the negative of the DVM and one hand.
don't forget to check idle current of the output transistors after it works okay. 20-100 ma is a normal range, with 20-40 more common.
I've fixed several amps with a DVM, an analog AC voltmeter with a 2 VAC scale, and a radio tuned to a rock station. It is not that hard. The analog meter will show beats of rock music if you are having trouble with the AC signal in the low voltage input circuitry (pre). I had a lot of transistors, diodes, resistors, and even 50 v capacitors blown open on the last amp I fixed. No scope, no capacitor meter, just a DVM. Most problems I found because the DC voltages were stupid; capacitors not blocking, diodes open or shorted etc. I just had to do a little AC signal analysis at the end to find a bad solder joint installed by the factory in 1997, that caused one channel to whang into DC out on the speakers if you looked that area cross-eyed. Note AC out with no signal should be zero, too, and if you have some, you have an oscillation you can't hear. You can prove it with the analog Ac meter with smaller and smaller capacitors in series- a RF signal will blow right through a 100 pf capacitor.
So have fun.
The reason those of us with experience change >15 year old electrolytic capacitors is that they are wet systems sealed with rubber. Historic rubber is dirt. Just because the device wasn't used, doesn't mean oxygen wasn't attacking the rubber seal on electrolytic capacitors every day since production. how good are the tires on a car made the year of you amp? The air pressure tests okay. (like a good 2v capacitance check). Want to take those tires out at full speed on the motorway/freeway/autobahn? Nah, that would be stupid. Same with old electrolytic capacitors, they can be fine today and slime all over the PWB tomorrow. It is better for amateur solderers to get the amp working, then change the capacitors for reliability one at a time, to focus your eyes on exactly the mistake you just made soldering it in.
Output transistors are the root of a lot of failure in small band amps due to operating into shorted speaker wires, especially the 1/4" phone plug pulled partway out. Amps in organs or fixed wiring situations like churches or theaters, don't nearly have as many shorted output transistors.
The speaker relay contacts have to be jumpered, and the DC protection circuitry (if any) has to be defeated to determine if you have DC output on the speaker terminals. the right answer is less than 50 mvdc. Than put the protection back before final watt testing the amp and closing it up.
Diode tests on DVM's check the junction on a transistor at low voltage and low current. Forwards voltage drop on a good junction can be from .45v to .73v. (multiply by 1000 on my dvm, I use 2000 ohms scale). Backwards drop is 1999 or dash. I find the diode drop test of a DVM will determine that a transistor or diode is bad, but that they can still blow up after they passed that test. Mooly etc likes to measure gain, I don't have equipment that sophisticated. I found a Vce leakage test correlates to old transistors that don't blow up. It requires a voltage source, I use a car battery charger with a capacitor on it to make 17 VDC. I short the B to E on the transistor, then put a milliammeter between the collector of the transistor under test and the battery charger. + on collector for NPN, - on collector for PNP. New transistors don't leak very much current, old used ones do sometimes. when I eliminated the leaky transistors they stopped blowing up in the amp after 30 seconds to 2 minutes. You could do this test at 150 V or 250 V or whatever the Vceo of the transistor is, but voltage over 24 is dangerous to you. The 17 v test seemed to screen out stressed used transistors.
I load the speaker terminals of the amp with 10 ohm 450 W resistors, instead of speakers, for test. I put a cheap car radio speaker protected from DC by a non-polar electrolytic capacitor, across 5 ohms of the output resistor. When intermittant problems occur in the amp, while it is playing music, the speaker sounds weird, without actually ripping out the voice coil from the DC out at the same time.
Don't test areas of the amp over 24 V using two hands, current across your heart can stop it. Use a clip lead on the negative of the DVM and one hand.
don't forget to check idle current of the output transistors after it works okay. 20-100 ma is a normal range, with 20-40 more common.
I've fixed several amps with a DVM, an analog AC voltmeter with a 2 VAC scale, and a radio tuned to a rock station. It is not that hard. The analog meter will show beats of rock music if you are having trouble with the AC signal in the low voltage input circuitry (pre). I had a lot of transistors, diodes, resistors, and even 50 v capacitors blown open on the last amp I fixed. No scope, no capacitor meter, just a DVM. Most problems I found because the DC voltages were stupid; capacitors not blocking, diodes open or shorted etc. I just had to do a little AC signal analysis at the end to find a bad solder joint installed by the factory in 1997, that caused one channel to whang into DC out on the speakers if you looked that area cross-eyed. Note AC out with no signal should be zero, too, and if you have some, you have an oscillation you can't hear. You can prove it with the analog Ac meter with smaller and smaller capacitors in series- a RF signal will blow right through a 100 pf capacitor.
So have fun.
Last edited:
That would come down to the specific device. The bulb drops the mains voltage seen by the unit, by how much depends on how much current is drawn. If the unit draw little current then the bulb would make no difference at all. Normally, we are talking of "linear" amplifier circuits (such as your amp) when we advise the bulb tester.
Mooly said 😛
Because measuring the B-E drop is a very useful test. The bulb won't alter the reading (unless the voltage to the amp dropped really low) because the transistors are really current driven devices. The B-E junction "clamps" the voltage to 0.7 volts irrespective of the current flowing.
The idea of the bulb is to help protect against failure of output transistors while working on an amp. You don't short speaker relays out, for most amps all that side of things would work normally as long as there is no fault on the amp causing it to draw excess current.
The Mains light bulb tester is very unlikely to protect a Live amp that gets accidentally shorted. The energy stored in the PSU will probably blow a semiconductor or many and maybe even some of the resistors.
Shorting a base to collector of an output, or driver, or pre-driver will almost certainly destroy some part of the amp, even when powered by a 40W test bulb.
The bulb tester is ideal for powered startup. The limiter action is on the point of cutting in. That is the "glow" we see. Drawing just a bit extra current changes that glow to a flash. It that slightly higher current is extended in time then the flash becomes continuous and the mains transformer then gets fed with mains + a big limiting resistor.
A bulb changes resistance from cold to hot by a factor of 6 to 10. That PTC action is very close to instability. It needs just a little kick and the cold becomes hot.
So Actually, when the receiver is ok, he will take low current , and because it's in series the bulb will also get low current? And the voltage will be split between the bulb and the amplifier (with the amp getting more of the 230 volt)
But if the receiver is shorted it works like a' ''wire''and then the bulb can take all the current it ones, lighting up and taking all the voltage?
So it's a combination of voltage and current(Ampere)
And a bigger amp needs a bigger bulb?
Edit:
On this site:
http://www.vintage-radio.com/projects/lamp-limiter.html
They say:
But at first they are talking about 400mA and later about 150mA But in series the current in amperes is equal in whole the arrangement.
I also don't really get what happens when the amp is short circuit? It behaves like wire, so all the current is going trough it, only the voltage doesn't drop?
Last edited:
You'll see how it all works for real on your amp because that is the way we will work on it to test it. Even turning the bias setting up a little will cause the lamp to glow more as the output transistors draw that bit more current.
A bigger amp doesn't automatically mean a bigger bulb because even a 100 or 200 watt amp may only draw a similar current to yours when it is idle.
A bigger amp doesn't automatically mean a bigger bulb because even a 100 or 200 watt amp may only draw a similar current to yours when it is idle.
About the bigger bulb for bigger amps.
Yeah, an amp with 20 output transistors and fixed voltage for op amps seems to need a 1500 w room heater instead of a light bulb on the AC in, or the op amps dont have enough voltage to work.
The tech or diyourselfer working in a shop on a flakey amp should guarentee with his ohmmeter that the speaker terminals are not shorted before he powers the amp up, so the "accidental shorted amp" Andrew T quoted has nothing to do with the light bulb tester. What the light bulb does do is flash when the amp is internally shorted. (or a room heater makes crinkly noises when the amp shorts ). I had a half dozen used output and driver transistors that passed the diode test on a DVM at 2 volts, but some would explode violently after 30 seconds to 2 minutes of power on at 85 volts. As I said the current leakage test at 17 vdc seems to eliminate those, but having the light bulb or heater element in the amp AC input limits the velocity of the exploding semi dies if you didn't catch them all. The light bulb or heater in the AC line also may limit collateral damage to other parts (like emitter resistors) if a used semiconductor does short when it heats up.
The art of cheap diy repair is replacing everything that is failed or likely to fail soon, without replacing every **** part in the amp. The light bulb box in the AC line helps you through that process to identify the parts to replace.
I don't like mains power flying around on wires that came out from under the screw on the light bulb base, so I spent an entire weekend building a light bulb box with a power switch, a circuit breaker, a socket for 3 pin power plugs, a strain relieved power cord, and a grounded ventilated case. Then it didn't work because my amp was too big for a 100 W bulb. But when I walk around the coffee table with the amp project getting the mail or playing my hammond organ, there is no risk of hitting mains power or breaking a light bulb. Suggest anybody using a light bulb box build the same arrangement.
Yeah, an amp with 20 output transistors and fixed voltage for op amps seems to need a 1500 w room heater instead of a light bulb on the AC in, or the op amps dont have enough voltage to work.
The tech or diyourselfer working in a shop on a flakey amp should guarentee with his ohmmeter that the speaker terminals are not shorted before he powers the amp up, so the "accidental shorted amp" Andrew T quoted has nothing to do with the light bulb tester. What the light bulb does do is flash when the amp is internally shorted. (or a room heater makes crinkly noises when the amp shorts ). I had a half dozen used output and driver transistors that passed the diode test on a DVM at 2 volts, but some would explode violently after 30 seconds to 2 minutes of power on at 85 volts. As I said the current leakage test at 17 vdc seems to eliminate those, but having the light bulb or heater element in the amp AC input limits the velocity of the exploding semi dies if you didn't catch them all. The light bulb or heater in the AC line also may limit collateral damage to other parts (like emitter resistors) if a used semiconductor does short when it heats up.
The art of cheap diy repair is replacing everything that is failed or likely to fail soon, without replacing every **** part in the amp. The light bulb box in the AC line helps you through that process to identify the parts to replace.
I don't like mains power flying around on wires that came out from under the screw on the light bulb base, so I spent an entire weekend building a light bulb box with a power switch, a circuit breaker, a socket for 3 pin power plugs, a strain relieved power cord, and a grounded ventilated case. Then it didn't work because my amp was too big for a 100 W bulb. But when I walk around the coffee table with the amp project getting the mail or playing my hammond organ, there is no risk of hitting mains power or breaking a light bulb. Suggest anybody using a light bulb box build the same arrangement.
Last edited:
well ..it seems that you are looking for a freebie tele repair .... Eventhough members will have the intention to share opinions and knowledge this will be in a given language which i am not sure that you understand properly .
You are asking specific repair tips but the proper way to implement those especially for a diyer is to start from step A to step B and so on .
One option can be to follow with eyes closed the already given tips and do that in a no questions asked level while the other option can be to ask as many questions as you like but work as a diyer start from small things and get to the next level with one step at a time approach ...
It will consume ages to proper understand all the nice help given here ...
Kind regards
Sakis
You are asking specific repair tips but the proper way to implement those especially for a diyer is to start from step A to step B and so on .
One option can be to follow with eyes closed the already given tips and do that in a no questions asked level while the other option can be to ask as many questions as you like but work as a diyer start from small things and get to the next level with one step at a time approach ...
It will consume ages to proper understand all the nice help given here ...
Kind regards
Sakis
Well, i am not looking for a freebie repair... It's more like I wan to learn it myself. It doesn't matter it takes ages,but I find it very interesting.
Next year I am going to do electrical engineering so I hope everything gets a bit clearer then.
It would also be nice if there are some theoretical books about amplifiers, but they are hard to find..
The thing is that at this moment I a somebody that is able to replace some capacitors/ replace a fuse, and clean potmeters. But I don'tfully understand the things behind it.
But for example: If there's something wrong some people say: It are bad caps, or: you Always need to replace capacitors after 25 years. But I am just curious if that's really true.
EDIT:
I want to recap my rotel ra-1412 when it's back from repairs (problems that were too difficult for me to find) but do I need to replace the Polypropylene film caps for example?
I thought these were capacitors that could last up to more then 60 years?
Next year I am going to do electrical engineering so I hope everything gets a bit clearer then.
It would also be nice if there are some theoretical books about amplifiers, but they are hard to find..
The thing is that at this moment I a somebody that is able to replace some capacitors/ replace a fuse, and clean potmeters. But I don'tfully understand the things behind it.
But for example: If there's something wrong some people say: It are bad caps, or: you Always need to replace capacitors after 25 years. But I am just curious if that's really true.
EDIT:
I want to recap my rotel ra-1412 when it's back from repairs (problems that were too difficult for me to find) but do I need to replace the Polypropylene film caps for example?
I thought these were capacitors that could last up to more then 60 years?
Last edited:
I read this topic on audiokarma:
Symptoms and Repair of Bad Caps. - AudioKarma.org Home Audio Stereo Discussion Forums
So I think the main conclusion is: Caps go bad, but more often some other things are the cause of problems. Replacing is not Always necesarry, and if it's your hobby just do the thing.
But what they don't answer is: what if a big (or a small) cap is really bad. What does it causes? Bad sound, or no sound? Or more damage.
Like if I get a amplifier that reacts good on the bulb test but sounds bad, when is it possible it's a cap?
Symptoms and Repair of Bad Caps. - AudioKarma.org Home Audio Stereo Discussion Forums
So I think the main conclusion is: Caps go bad, but more often some other things are the cause of problems. Replacing is not Always necesarry, and if it's your hobby just do the thing.
But what they don't answer is: what if a big (or a small) cap is really bad. What does it causes? Bad sound, or no sound? Or more damage.
Like if I get a amplifier that reacts good on the bulb test but sounds bad, when is it possible it's a cap?
Electrolytic is a word that is spelled differently than the word, polyprophylene. Thus electrolytic caps are not polyprophylene caps. The plastic caps I own seem to have to life limit. But they were only produced in my gear after 1966.
electrolytic caps have a plus on one end, a stripe with a minus in balls pointing to one lead, or a voltage followed by the letters "NP" for non polar.
Guitar amps get some shorted plastic film caps because 1. they are banged around every night setting up and tearing down and 2. many guitar amp builders use cheap caps. So guitar tube amp repairers are focussed on changing every high voltage cap. I don't have that problem in the hammond organs or the Dynaco equipment, I just leave the plastic (and dynaco paper) caps be. If one shorts it will have a stupid DC voltage reading.
Shorted e-caps near the power transformer or main rectifier on a switcher, shorts out the rectifier, a resistor, the transformer, or takes the fuse out. High or low value e-caps (mostly) in a filter circuit make the filter point too high or too low, making the "hifi" or organ sound weird. I own three $5000 new in 1967-8 organs that sound like kazoos at 40 years of age because Hammond used over 200 electrolytic caps in them. I paid $200 for one, put 71 e caps in it, and it sounds like $5000 again. Now I'm offered three more for free.
Read about the vector impedance equation for RLC circuits and the inflection point of RLC filters if you're interested. Or wait until 2nd year physics, you'll take it then after you've had more vector algebra.
I've repaired two solid state amps and the first needed 3 output transistors, 3 driver transistors, two emitter and four bias resistors, a power regulator transistor, a zener diode and very little else. I replaced the e-caps based on the age (1970). The second solid state amp needed 19 output transistors, 3 driver transistors, 4 predriver transistors and 2 VI limiter transistors, 6 IC's, >50 diodes, >50 resistors, >10 ceramic and film capacitors, 4 small exploded e-caps, 5 burnt off PCB lands. When 170 v gets loose in a system due to blown output transistors it rips up stuff. Hope your next project is not that hard. I replaced about 30 1994 age e-caps in that one, after most of the repairs.
Oh, and the tube amp needed e-caps about every 8 years, and the tube preamp needed e-caps about every 12 years for power supply weakness. In the 43 years I've owned them. those did need a few rectifier and output tubes, unlike the Hammond organs. They also needed volume pots and power switches a few times.
Last edited:
- Status
- Not open for further replies.