One of my ESI nEar 05 studio monitors has recently stopped working. I noticed it wasn't working a few days after a 5min REALLY loud playing session and I'm thinking that I may have blown it, although it might just be coincidence.
The symptoms are that it switches on OK, but the only output is a fairly quiet mid-freq buzz that you can't hear until your ear is about a foot away from the speaker. Adjusting the master volume pot makes no difference to the level of the buzz. The other monitor still works fine.
I'm now armed with an oscilloscope, multimeter and some limited electronics knowledge and I'm hoping to debug the problem. I can obviously use the good speaker as a reference if need be.
I've noticed that the internal electronics are fairly simple - there are only 3x ICs on the board - 2x LM3886TF power amps and 1x unknown 8-pin IC marked 4558D JRC 45E0311.
I was thinking that I'd maybe just replace the 2x LM3886TFs and hope for the best. If I knew what the 8-pin IC was too I'd replace that. Other than this, I could try debugging it with the scope, the good monitor and an audio source, but I wouldn't know where to start since I don't have a schematic.
For those that are interested, here are some photos:
http://i800.photobucket.com/albums/yy282/stube40/IMG_0232Large.jpg
http://i800.photobucket.com/albums/yy282/stube40/IMG_0231Large.jpg
http://i800.photobucket.com/albums/yy282/stube40/IMG_0230Large.jpg
http://i800.photobucket.com/albums/yy282/stube40/IMG_0229Large.jpg
http://i800.photobucket.com/albums/yy282/stube40/IMG_0233Large.jpg
The symptoms are that it switches on OK, but the only output is a fairly quiet mid-freq buzz that you can't hear until your ear is about a foot away from the speaker. Adjusting the master volume pot makes no difference to the level of the buzz. The other monitor still works fine.
I'm now armed with an oscilloscope, multimeter and some limited electronics knowledge and I'm hoping to debug the problem. I can obviously use the good speaker as a reference if need be.
I've noticed that the internal electronics are fairly simple - there are only 3x ICs on the board - 2x LM3886TF power amps and 1x unknown 8-pin IC marked 4558D JRC 45E0311.
I was thinking that I'd maybe just replace the 2x LM3886TFs and hope for the best. If I knew what the 8-pin IC was too I'd replace that. Other than this, I could try debugging it with the scope, the good monitor and an audio source, but I wouldn't know where to start since I don't have a schematic.
For those that are interested, here are some photos:
http://i800.photobucket.com/albums/yy282/stube40/IMG_0232Large.jpg
http://i800.photobucket.com/albums/yy282/stube40/IMG_0231Large.jpg
http://i800.photobucket.com/albums/yy282/stube40/IMG_0230Large.jpg
http://i800.photobucket.com/albums/yy282/stube40/IMG_0229Large.jpg
http://i800.photobucket.com/albums/yy282/stube40/IMG_0233Large.jpg
The 4558D is a very common inexpensive audio op-amp made by new japan radio corp amongst others.. Data sheet here: http://semicon.njr.co.jp/njr/hp/fileDownloadMedia.do?_mediaId=7903
OK, so you have a deep-fried self powered 2 or 3 way mini-monitor? First check that there is no DC on the speaker outputs to burn out the drivers.
Then check that you do have DC fom the power supply by measuring its output from the power electrolytics and does this appear again at the power Chip-amps (LM3886)? Take real care not to short pins here - one short and its bucks + an awful job refitting if you aren't a tech. or experienced repairer.
Then check that you do have DC fom the power supply by measuring its output from the power electrolytics and does this appear again at the power Chip-amps (LM3886)? Take real care not to short pins here - one short and its bucks + an awful job refitting if you aren't a tech. or experienced repairer.
Hi Ian,
Thanks for the advice.
I've ordered new LM3886 and 4558D chips from Digikey ($7 and $0.50 each) Will be here in a few days. I've ordered a few spares too in case I blow some up.
I see what you mean about fitting them, but I think I'll manage OK. Would you recommend I just go ahead and fit them or would it be worth me trying to debug as per your instructions first of all?
Stuart.
Hi Stuart
Without some feedback on the basic condition there, it won't help to guess what next. With no schematic either, we fly blind and hope to collide with the problem. However, I would be surprised if the power chips failed - they are almost bulletproof for the very reason you are in this predicament - you know....turn it up until....etc.
Make sure the power supply is working first and check the output for DC. If this is not easily determined, you might reconsider the sense of just replacing the semiconductors in the hope that this will solve the problems.
Consider this is a set of amplifiers at different power levels driving separate speakers and somehow all 3 amps have failed? You could replace them and immediately farewell your new parts the moment you power up too, if the power supply is badly at fault.
The 4558 dual opamp is a very low power device, a preamp or more likely crossover filter
gain block to split the signal to the amps. Check its power rails which may be +/- 15V at pins 4,8 with OV ( Power ground) as common. Check whether the power supply is a dual rail or perhaps single with isolating caps. at the outputs. (this is often done for low cost)
In that case you may only have a single power rail to th 3886s of somewhere around 40V.
NB: need to know is this a 2 way or 3 way as 2 way config. may be bridged amps. Nasty!
Being actually a Chipamp matter, it might get more suitable attention in that forum.
Without some feedback on the basic condition there, it won't help to guess what next. With no schematic either, we fly blind and hope to collide with the problem. However, I would be surprised if the power chips failed - they are almost bulletproof for the very reason you are in this predicament - you know....turn it up until....etc.
Make sure the power supply is working first and check the output for DC. If this is not easily determined, you might reconsider the sense of just replacing the semiconductors in the hope that this will solve the problems.
Consider this is a set of amplifiers at different power levels driving separate speakers and somehow all 3 amps have failed? You could replace them and immediately farewell your new parts the moment you power up too, if the power supply is badly at fault.
The 4558 dual opamp is a very low power device, a preamp or more likely crossover filter
gain block to split the signal to the amps. Check its power rails which may be +/- 15V at pins 4,8 with OV ( Power ground) as common. Check whether the power supply is a dual rail or perhaps single with isolating caps. at the outputs. (this is often done for low cost)
In that case you may only have a single power rail to th 3886s of somewhere around 40V.
NB: need to know is this a 2 way or 3 way as 2 way config. may be bridged amps. Nasty!
Being actually a Chipamp matter, it might get more suitable attention in that forum.
Last edited:
Thanks Ian - I think I'll take your advice and take a look at the power rails whilst waiting for the components to arrive.
I don't quite understand your question about it being a 2-way, or a 3-way as a 2-way etc. I'm not very well-versed in audio jargon and I guess you are talking about the speaker config? If so, it has a single tweeter and a single woofer - both tweeter and woofer have 2 wires each connecting to the main circuit board but I haven't had a look to see what's going on there in terms of crossover configuration. Does this help, or could I investigate further to get you some more info?
I don't quite understand your question about it being a 2-way, or a 3-way as a 2-way etc. I'm not very well-versed in audio jargon and I guess you are talking about the speaker config? If so, it has a single tweeter and a single woofer - both tweeter and woofer have 2 wires each connecting to the main circuit board but I haven't had a look to see what's going on there in terms of crossover configuration. Does this help, or could I investigate further to get you some more info?
Thanks Stuart, for the confirmation. 2-way means 2 speakers here. These are active speakers, so each driver has its own amplifier and the crossover is done at signal level actively (in the preamp circuit) by an opamp. So treble then goes to the tweeter amp and bass to the Mid/bass amp and then direct to their respective speaker drivers. Conventional passive speaker crossover coils and caps. are avoided in this arrangement.
Since you have located the speaker leads, test with a Digital Multimeter for DC across them. I assume you have basic tools to do this or you would not be trying to DIY electronic repairs. There should not be much more than 40 mV DC at the outputs.
Since there are 3 power amplifier chips and only 2 speakers, it's likely the bass amplifier is configured as a bridged pair of 3886s and things now get complicated. So, is there still output of scratchy sound from the tweeter when powered on with programme playing or is that dead too? Verify that the speakers are functional by testing the resistance across their terminals - disconnected from the amplifier! It should read a bit less than the marked impedance and a little scratchy sound should be evident when you test. If you cannot tell if sound is coming from the tweeter, disconnect the bass speaker so you can.
It is important to logically check that everything else works before wasting time and money on guesswork. Might I suggest you read the article pages on Rod Elliott's ESP site to get an idea of how active speakers, chip amps and bridge configurations are done and operate. Elliott Sound Products - The Audio Pages (Main Index)
It's no cakewalk fixing broken gear without some basic knowledge of how it works and how to check that, at least. Good luck - it may be simpler and easier after all!
Since you have located the speaker leads, test with a Digital Multimeter for DC across them. I assume you have basic tools to do this or you would not be trying to DIY electronic repairs. There should not be much more than 40 mV DC at the outputs.
Since there are 3 power amplifier chips and only 2 speakers, it's likely the bass amplifier is configured as a bridged pair of 3886s and things now get complicated. So, is there still output of scratchy sound from the tweeter when powered on with programme playing or is that dead too? Verify that the speakers are functional by testing the resistance across their terminals - disconnected from the amplifier! It should read a bit less than the marked impedance and a little scratchy sound should be evident when you test. If you cannot tell if sound is coming from the tweeter, disconnect the bass speaker so you can.
It is important to logically check that everything else works before wasting time and money on guesswork. Might I suggest you read the article pages on Rod Elliott's ESP site to get an idea of how active speakers, chip amps and bridge configurations are done and operate. Elliott Sound Products - The Audio Pages (Main Index)
It's no cakewalk fixing broken gear without some basic knowledge of how it works and how to check that, at least. Good luck - it may be simpler and easier after all!
Wow, I certainly came to the right place!! When I first posted my cry for help, my utopian scenario was that some helpful individual would post information like this. However, I didn't actually believe that would happen - but thanks to you it has!!
I have both a scope and multimeter available to me (I work in a company that has an electronics lab that I have access to and occasionaly use).
With your information and the new components, I'm pretty sure I'll get to the bottom of this. I'll obviously write up the findings so that other people can benefit from it.
I have both a scope and multimeter available to me (I work in a company that has an electronics lab that I have access to and occasionaly use).
With your information and the new components, I'm pretty sure I'll get to the bottom of this. I'll obviously write up the findings so that other people can benefit from it.
OK, so changed the 2x LM amps and the little JRC amp but the problem has not gone away. Still get a fixed mid-range freq buzz coming from the speaker when I turn it on. The buzz is still fairly quiet, although seems a fraction louder now than before. The buss is still unaffected by the position of the master volume pot.
I guess I will have to debug it now based on Ian's previous instructions. Will report back soon.
I guess I will have to debug it now based on Ian's previous instructions. Will report back soon.
I think it might be the bridge rectifier. It's a RS406 and I notice that it has a clean 24V coming out of the +ve terminal, but the -ve terminal has a nasty 100Hz distorted sine wave on it with 20V amplitude (actually, it's like a hybrid between a sine wave and a triangle wave).
I guess this means that a diode in the bridge rectifier is gone?
Next thing is to try and find somewhere that sells the RS406.
Or, could I make my own RS406 up from 4 diodes?
I guess this means that a diode in the bridge rectifier is gone?
Next thing is to try and find somewhere that sells the RS406.
Or, could I make my own RS406 up from 4 diodes?
Hi Stuart, Yes, I think you are closer to the mark here. Measure the DC voltages. Its nice to know that you don't have sinewaves but they won't look much better after rectification without a load and decent electrolytic capacitors either. Assuming the diode bridge is blown (and being only 4 amp max, type from the pics,) you can get replacements everywhere. Altronics #Z0078 or Z0079, depending whether pin spacing is 6 or 7 mm should do. Don't use higher rated parts, unless you actually want a worse disaster next time. It would seem that proper protective fuses are not fitted or the wrong rating. That's not surprising with budget gear, but it still should be corrected so that fuses and not parts fail.
Look at the voltage rating of the largest electrolytics. This will be higher than the DC you expect across them. Let's say it reads 35V, so you may expect around 30V DC across the terminals. If not, you certainly have a problem.
Be methodical; when the bridge is removed , check the transformer wiring to it. You may have a single (2 wires, terminals etc.) or double windings (4 terminals with the windings in series and the common terminal earthed. Either way, the windings must have AC output on them when disconnected from the diode bridge. If not, turn off, remove power and test the transformer primary (240V connection) for continuity. You should read around 30 ohms here.
Assuming the transformer is OK for continuity, replace the bridge but you need to protect the system with a very light fuse on start up as there still could be bad faults like shorted or open circuit electrolytics. With this kind of stress, you could well have dangerous, explosive devices there - you have been warned, take personal safety precautions when first powering up after a bridge fault.
Please post voltage readings as indicated. If you can't be certain, just say so. Thanks
Look at the voltage rating of the largest electrolytics. This will be higher than the DC you expect across them. Let's say it reads 35V, so you may expect around 30V DC across the terminals. If not, you certainly have a problem.
Be methodical; when the bridge is removed , check the transformer wiring to it. You may have a single (2 wires, terminals etc.) or double windings (4 terminals with the windings in series and the common terminal earthed. Either way, the windings must have AC output on them when disconnected from the diode bridge. If not, turn off, remove power and test the transformer primary (240V connection) for continuity. You should read around 30 ohms here.
Assuming the transformer is OK for continuity, replace the bridge but you need to protect the system with a very light fuse on start up as there still could be bad faults like shorted or open circuit electrolytics. With this kind of stress, you could well have dangerous, explosive devices there - you have been warned, take personal safety precautions when first powering up after a bridge fault.
Please post voltage readings as indicated. If you can't be certain, just say so. Thanks
OK, the new bridge-rectifier didn't fix the problem at all.
So, I got suspicious of the electrolytic caps on the output of the bridge. Using a multimeter I noticed about 1Mohm across the cap on the +ve rail (the good rail), but only about a 1/4th of this on the cap on the -ve rail. However, who knows what else in the circuit is in parallel with this.
Anyway, I removed the -ve rail cap and put it on an LCR meter. The LCR quoted a figure of 86nF for the cap value but the cap is 6800uF. Pulling a 1000uF cap off the shelf to test the LCR, it quoted 800uF - so proving the LCR is working reasonably well.
Hence, looks like I need to replace the mains filtering cap. It's a 6800uF 25V made by Acon. It's about 1cm in dia by about 4cm long. Searching on farnell brings up a whole load of different ones and I'm not sure what to choose.
Does it come down to wattage?
If so, the speaker quotes the LF amp as having a power rating of 40W, with the HF amp being 30W. Hence, I assume that means the whole powered speaker is consuming less than 100W??
So, I got suspicious of the electrolytic caps on the output of the bridge. Using a multimeter I noticed about 1Mohm across the cap on the +ve rail (the good rail), but only about a 1/4th of this on the cap on the -ve rail. However, who knows what else in the circuit is in parallel with this.
Anyway, I removed the -ve rail cap and put it on an LCR meter. The LCR quoted a figure of 86nF for the cap value but the cap is 6800uF. Pulling a 1000uF cap off the shelf to test the LCR, it quoted 800uF - so proving the LCR is working reasonably well.
Hence, looks like I need to replace the mains filtering cap. It's a 6800uF 25V made by Acon. It's about 1cm in dia by about 4cm long. Searching on farnell brings up a whole load of different ones and I'm not sure what to choose.
Does it come down to wattage?
If so, the speaker quotes the LF amp as having a power rating of 40W, with the HF amp being 30W. Hence, I assume that means the whole powered speaker is consuming less than 100W??
Hi Stuart
Indication is that that you have average 24V DC with 25V rated caps. This is not a safe margin and will often be exceeded with power surges. You don't need to order special parts. Just measure the pin spacing so new ones fit and buy an electro with preferably a higher, say 35V rating, from your nearest Jaycar, Altronics, Rockby or any electronic parts store. If the original is glued to the board for security, use contact cement or silastic on the replacement too.
There are no power ratings on caps. Just working voltage and capacitance. When devices are hot, temperature ratings come into play but here, in a timber cabinet, I don't think so. You could use any cap between 5,600 and 10,000 uF here, if you couldn't readily obtain 6800 uF. It would be nice to have an industry standard like Panasonic brand but the retailer's brands are quite acceptable.
Have you measured the AC voltage from the transformer yet and did the diode measurements of the bridge show fault? As Michael asks, have you found fuses yet?
Indication is that that you have average 24V DC with 25V rated caps. This is not a safe margin and will often be exceeded with power surges. You don't need to order special parts. Just measure the pin spacing so new ones fit and buy an electro with preferably a higher, say 35V rating, from your nearest Jaycar, Altronics, Rockby or any electronic parts store. If the original is glued to the board for security, use contact cement or silastic on the replacement too.
There are no power ratings on caps. Just working voltage and capacitance. When devices are hot, temperature ratings come into play but here, in a timber cabinet, I don't think so. You could use any cap between 5,600 and 10,000 uF here, if you couldn't readily obtain 6800 uF. It would be nice to have an industry standard like Panasonic brand but the retailer's brands are quite acceptable.
Have you measured the AC voltage from the transformer yet and did the diode measurements of the bridge show fault? As Michael asks, have you found fuses yet?
The fuses were OK.
I replaced the 6800uF cap with a 4700uF cap and it works - I couldn't get hold of a 6800uF cap anywhere locally. I knew I had fixed the problem it before I even put audio through it since the scope showed that the -ve rail was nice and flat and the old waveform had disappeared.
However, there is a slight buzz (like an earth loop buzz) coming from the speaker still although this time the buzz has a volume that changes with the master pot and it disappears altogether at zero volume. Other than that, it plays music fine now showing that it is indeed fixed and that I've fitted the new amp ICs and bridge properly.
I'm wondering if the slight buzz is related to an inbalance in the ripple on both -ve and +ve rails now that the -ve has a smaller cap on it. So, I've ordered 2 new Panasonic 6800uF ones from Farnell rated to 35V (pt no 1198693). I thought it would be better to replace them as a pair.
I replaced the 6800uF cap with a 4700uF cap and it works - I couldn't get hold of a 6800uF cap anywhere locally. I knew I had fixed the problem it before I even put audio through it since the scope showed that the -ve rail was nice and flat and the old waveform had disappeared.
However, there is a slight buzz (like an earth loop buzz) coming from the speaker still although this time the buzz has a volume that changes with the master pot and it disappears altogether at zero volume. Other than that, it plays music fine now showing that it is indeed fixed and that I've fitted the new amp ICs and bridge properly.
I'm wondering if the slight buzz is related to an inbalance in the ripple on both -ve and +ve rails now that the -ve has a smaller cap on it. So, I've ordered 2 new Panasonic 6800uF ones from Farnell rated to 35V (pt no 1198693). I thought it would be better to replace them as a pair.
Ok, progress.
Without a significant signal and hence load on the power supply, even a few hundred uF will give smooth DC and guiet output. If the buzz remains, it's possible other caps are blown too. Check that the buzz comes from both or only the bass speaker as previously described. If only from that bass amp, then there is the circuit around the 2 x 3886 chips so check for more failed caps, open diodes, suspect looking resistors. by using your DMM. Don't be fooled by a lot o readings which will unavoidably include other parallel and series parts. Look around the 4558 parts and it's supply to pins 4 & 8 too.
There is a fault, even if just an unnoticed solder bridge or unmade connection when refitting the chips. Keep going with a light and careful touch - you are so close.
Without a significant signal and hence load on the power supply, even a few hundred uF will give smooth DC and guiet output. If the buzz remains, it's possible other caps are blown too. Check that the buzz comes from both or only the bass speaker as previously described. If only from that bass amp, then there is the circuit around the 2 x 3886 chips so check for more failed caps, open diodes, suspect looking resistors. by using your DMM. Don't be fooled by a lot o readings which will unavoidably include other parallel and series parts. Look around the 4558 parts and it's supply to pins 4 & 8 too.
There is a fault, even if just an unnoticed solder bridge or unmade connection when refitting the chips. Keep going with a light and careful touch - you are so close.
OK - replaced both caps on the output of the bridge with shiny new identical Panasonic 35V 6800uF caps. The buzz is totally gone now.
The only possible problem is related to the size of the caps which are bigger since they are rated to 35V instead of 25V. This means they cannot be made flush with the PCB and are lifted above it by about 1cm. This might create an issue of rattling with the audio vibrations. I've tried to secure them with some silicone sealant, but they still have the ability to rattle. We'll see how they perform in the long run...........
So, it seems we reached the end of the road. In the end it was a very simple fix to the problem, the most difficult part was all the mucking around to find out what the problem actually was.
Thanks for everyone's help especially Ian.
The only possible problem is related to the size of the caps which are bigger since they are rated to 35V instead of 25V. This means they cannot be made flush with the PCB and are lifted above it by about 1cm. This might create an issue of rattling with the audio vibrations. I've tried to secure them with some silicone sealant, but they still have the ability to rattle. We'll see how they perform in the long run...........
So, it seems we reached the end of the road. In the end it was a very simple fix to the problem, the most difficult part was all the mucking around to find out what the problem actually was.
Thanks for everyone's help especially Ian.
Great result.
The physical size is part of buying stuff from the big suppliers. It's certainly unfortunate though, that you now have this problem of unsafe mounting.
I guess you know how much handling little near-field monitors usually get, so perhaps you could make up small round spacers from insulating material - even dense wood like dowelling rod that is close to the same diameter of the new caps but chiselled to fit neatly in the space on the PCB.
Then drill it to allow just the leads to pass through without stress. Then bond that, sandwiched between board and cap. It can't hurt and with occasional inspection, will be an improvement to mechanical security.
Alternatively, if you can post the pin spacing and diameter of the original caps, perhaps we can locate a suitable size from another source.
The physical size is part of buying stuff from the big suppliers. It's certainly unfortunate though, that you now have this problem of unsafe mounting.
I guess you know how much handling little near-field monitors usually get, so perhaps you could make up small round spacers from insulating material - even dense wood like dowelling rod that is close to the same diameter of the new caps but chiselled to fit neatly in the space on the PCB.
Then drill it to allow just the leads to pass through without stress. Then bond that, sandwiched between board and cap. It can't hurt and with occasional inspection, will be an improvement to mechanical security.
Alternatively, if you can post the pin spacing and diameter of the original caps, perhaps we can locate a suitable size from another source.
Last edited:
Hi Ian - I think you're right - one day I'll desolder the caps and use some home made spacer. Probably a rubber-bushing material with holes cut into them bonded to the caps via silicone or some other flexible adhesive.
First of all, I'll use them and judge for myself sonically whether the effort is worth it or not...........
First of all, I'll use them and judge for myself sonically whether the effort is worth it or not...........
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.