He will post information here, hopefully someone can help him fix his scope O0
Quote from: Peterae on 2013.11.26, 16:08:58
He will post information here, hopefully someone can help him fix his scope O0
Thank you Peter.
I have managed to "Abuse", accidentally BTW, my Fluke PM 3092. I managed to discharge a 50 micro farad capacitor with about 350 Vdc onto the earth clip!! This took out the 5 A fuse and upon replacement the scope comes on for a few minutes then the trace disappears coinciding with a high pitched whistle from the PSU transformer.
I now have the scope bared and I have also noticed that upon applying the probe to the signal ref pin (on the front) Ch1 shows a very deformed square wave whilst the other three channels appear to be ok.
I am open to any help!! Please ask whatever, I can even make a You tube if that is any help?
Cheers Grum.
If it has a dc-dc psu, then thats the squeal. Being that it happens after time on, most likely it is a capacitor that is heating and then internally shorting, thus the squeal of the psu being heavily loaded. It seems like a heat up problem. You can try using what is called freeze spray or a can of Blast Away duster turned upside down and cool caps in the circuit one at a time to see if the issue goes away temporarily. If you find a cap that reacts to cooling, then thats your bugger. ;) It could be a semiconductor, but I would go after the caps first. ;)
Mags
Dear Mags.
Thanks for that I will give it a try!!
Cheers Grum.
Quote:
QuoteI have managed to "Abuse", accidentally BTW, my Fluke PM 3092. I managed to discharge a 50 micro farad capacitor with about 350 Vdc onto the earth clip!! This took out the 5 A fuse and upon replacement the scope comes on for a few minutes then the trace disappears coinciding with a high pitched whistle from the PSU transformer.
I now have the scope bared and I have also noticed that upon applying the probe to the signal ref pin (on the front) Ch1 shows a very deformed square wave whilst the other three channels appear to be ok.
A more precise explanation of what you did to cause the failure would help e.g
QuoteI managed to discharge a 50 micro farad capacitor with about 350 Vdc onto the earth clip!!
Two points are needed to form a discharge circuit. We have one, the ground clip. What was the other connection to the capacitor?
QuoteI now have the scope bared and I have also noticed that upon applying the probe to the signal ref pin (on the front) Ch1 shows a very deformed square wave whilst the other three channels appear to be ok.
[/quote]
This is cryptic. Are you measuring the reference pin with a different scope? Are three channels of the scope operating properly before fadeout? What does "ok" mean. (no deformed square wave? no wave display at all?)
If three channels are operating "ok" as you said, part of the the problem is in the bad channels vertical amp.
Please try to re-explain step by step what you did and what you now have. Identify if other pieces of equipment are being used to test the faulty scope.
You can learn a lot in the few minutes the scope does work before it fades out. The fadeout is most likely a power supply problem, without a schematic it may be difficult to repair except by component swap i.e. the switchmode controller chip, drive fet, transistor or shorted turn in PS transformer.
Procure a service manual with schematics as a first step. It will be handy to have for this and future reference.
Dear ION.
Thank you for your response, Chet phoned me and said you would be in touch.
I PM'd you before I noticed your post!!
Please let me know if you need more info. BTW I now have a PDF of the service manual is there a way of attaching single pages?? The whole document is 120 MB!!
Cheers Grum.
Please let me know if you need more info. BTW I now have a PDF of the service manual is there a way of attaching single pages?? The whole document is 120 MB!!
Cheers Grum.
[/quote]
copy and paste to another file the pages you want, and then send or post here O0
regards
Mike
Thanks Mike!!
Attached schematic of voltage control.
C1012 was warmer than all the others for the same period of time.
The converter is a modified blocking oscillator design as TEK frequently uses with all the frills for current sensing and limiting, hence the squeal.
While it is possible there is a bad part in the power supply, another view is that it may be in CH1 loading the supply excessively.
Best to try and determine what component has failed and is causing such a load on the converter that it has altered it's duty cycle to provide strong harmonics in the audio range. This component may not be in the power supply itself, but the supply is reacting to the excessive loading by operating closer to 50% duty cycle and eventually shutting down.
The schematic indicates around a 25% duty cycle for the converter. This will be difficult to probe so don't even try it with another scope unless you know how to get around the ground loop problems you will have or more damage will ensue. Probe only the secondary isolated section.
After I download the manual I will see if Ch1 can be isolated or removed to see if the scope operates normally until the faulty component(s) can be found.
Ok Got the complete manual. Note of caution attached.
More to come once I give it a look.
Question: Was the probe tip of CH1 hooked to anything when the capacitor short occurred? If not, then the transient induced in the probe shield may have coupled into the inner conductor and taken out the input buffer amplifier. Since you get a distorted waveform on CH1, the buffer may not be operating properly and you are only getting pulse edges without the low frequency information.
See if the Ch1 Low frequency impedance converter IC type AD548 (N1001) is running much hotter than equivalent parts in CH2,3,4 (input attenuator board)
The fact that the signals are good on CH2,3,4 but distorted on Ch1 points to a defective part such as the AD548 in the attenuator section. This part gets hit first and hardest with a large transient on that channel and the shorted part may be loading the supply.
A careful check of startup power supply voltages especially +/-12V would be helpful.
Quote from: ION on 2013.11.27, 21:53:28
Ok Got the complete manual. Note of caution attached.
More to come once I give it a look.
Question: Was the probe tip of CH1 hooked to anything when the capacitor short occurred? If not, then the transient induced in the probe shield may have coupled into the inner conductor and taken out the input buffer amplifier. Since you get a distorted waveform on CH1, the buffer may not be operating properly and you are only getting pulse edges without the low frequency information.
See if the Ch1 Low frequency impedance converter IC type AD548 (N1001) is running much hotter than equivalent parts in CH2,3,4 (input attenuator board)
The fact that the signals are good on CH2,3,4 but distorted on Ch1 points to a defective part such as the AD548 in the attenuator section. This part gets hit first and hardest with a large transient on that channel and the shorted part may be loading the supply.
A careful check of startup power supply voltages especially +/-12V would be helpful.
Dear ION.
As far as I recall the probe tip was not connected when the accident occurred.
I have run up the scope with the main board's underside exposed and cannot detect any difference in temperature using the back of my finger as a guide. As it is getting late here I will have a look at the +/- 12 v area tomorrow. Can you suggest where I might find that test point??
Cheers Grum.
Quote from: Grumage on 2013.11.27, 23:17:15
Dear ION.
As far as I recall the probe tip was not connected when the accident occurred.
I have run up the scope with the main board's underside exposed and cannot detect any difference in temperature using the back of my finger as a guide. As it is getting late here I will have a look at the +/- 12 v area tomorrow. Can you suggest where I might find that test point??
Cheers Grum.
Grum:
The part I am talking about is on Ch1 of the vertical amplifier /Attenuator which is on a separate board where the probes plug into. Each of the channels has some power supply decoupling to prevent crosstalk.
Voltages can be measured on the large connector at the output of the power supply board. If you intend to look at the duty cycle, and frequency, do it on one of the isolated secondary windings for safety.
It will be a good idea to get really familiar with the power supply switcher circuit and how it operates. Read section 5.6-1 You will learn it has current sensing and limiting built in, so under heavy load from a shorted IC it will work somewhat, but will quit after some time.
Since Ch1 is not working during start up, we can have good reason to look for a shorted part there. The signal unit A1 board is not part of the main board, but is the board where the input channels plug into. Can you post a picture of ch1 hooked to the ref pin so we can determine the nature of the distorted signal? Also show a sample of Ch2 if possible.
The use of a BRY62 silicon controlled switch for current sensing / switch turnoff is interesting. I have used ordinary SCR's as current sense / turnoff for this type of switcher, but the use of the SCS is a new twist. Always learn something new from those TEK engineers. This is a well thought out flyback converter, far more elegant than a simple JT or blocking oscillator.
Attached is a clearer copy of the power supply printed to pdf from the original document. notice that it is only 225 kb compared to 348 for the .png, but when zoomed you can read the component values clearly.
Dear ION.
Please find attached photos as requested.
I have shown both the front and rear of the main board. As you can see the input sockets are directly on this , not a separate board.
Cheers Grum.
I nominate myself for the TWIT of the year 2013!!
They are all upside down!! But why ? They were all ok viewed on my PC!!
I shall rotate and repost!!.
Cheers TWIT.
Just tried with rotated images. Still the same!!
As I suspected the bad channel is only passing high frequency information. This points to the AD548 low frequency impedance converter IC as I stated earlier:
QuoteSee if the Ch1 Low frequency impedance converter IC type AD548 (N1001) is running much hotter than equivalent parts in CH2,3,4 (input attenuator board)
The fact that the signals are good on CH2,3,4 but distorted on Ch1 points to a defective part such as the AD548 in the attenuator section. This part gets hit first and hardest with a large transient on that channel and the shorted part may be loading the supply.
You are very lucky, a lot worse could have happened to this scope, like destruction of the uP IC or one of its peripherals. The fact that the functions seem to work is a sigh of relief.
The trick now is to do the repair without causing further damage, as this is very easy to do when probing around. You must be very patient, work slowly and very carefully especially if you are not familiar with repairing this type of equipment. One slip of a probe could short something causing chain reaction destruction of a number of chips rendering the scope a useless boat anchor.
I have done this in my early days and have learned the lesson.If it's not the IC AD548 loading the supply, then it could be a shorted decoupling Zener or other component in that section . Study the attenuator front end and you will see how the power is routed.
Edit: It is on the back side of the board(small component side) in the vicinity of ch1. see attachment
it is a part of the main board, not a separate board as I first thought.I have some thoughts on how we can disconnect power to that section to see if the short is there.
If you have a kill a watt meter you will probably find the input current to the scope is higher than its rating by a bit because of the extra power supply load. no need to re-post the picture, I get the idea.
Dear ION.
Strangely my Wattmeter reads 20 W under the rated Scope usage (BTW also re checked same using 40/60 W Incan bulbs)
I managed to get the Steel sheet cover removed that was soldered and again there is no difference in op temp between the 4 little chips.
I look forward to receiving your next instructions!! Is it holiday time for you??
Cheers Grum.
Hi Grum
Yes it is the holiday time here.
I'm surprised all 4 little chips read the same temp, so the problem could be the power source to the chip.
I will probably not be back on this until tomorrow, meanwhile, patience, we will slowly home in on the problem.
If you are in a rush then I'm not the one to help. I find it best to use simple logic and process of elimination to find the faulty part and to proceed cautiously.
Great that the power supply can start up and run for a bit. This narrows the possibilities a great deal.
We need to check the power supply voltages on the secondary and see which one is out of spec. or we can take a closer look at the power that runs the Ch1 AD548 to see if it is up to snuff. I will take a look at the schematic as time permits.
For future reference, it is a good idea to get an average power drain and note it on the back of the unit when it is running on spec and with the intensity set to a medium level. Lately I've been doing this as part of documentation for all my equipment.
Odd that you are drawing less but it may be because the supply is in current limit fold back mode.
Dear ION.
My heartfelt thanks to you taking time out during the holiday period!! Please do not worry I am a very patient type "slowly and surely wins the race" !! O0
Take your time and enjoy the day.
I'm off to bed soon!! Cheers Grum.
Quote from: Grumage on 2013.11.28, 22:00:15
Dear ION.
My heartfelt thanks to you taking time out during the holiday period!! Please do not worry I am a very patient type "slowly and surely wins the race" !! O0
Take your time and enjoy the day.
I'm off to bed soon!! Cheers Grum.
Dear Grumage:
After giving it more thought it is always possible that there is more than one failed part in the scope. We know for sure that low frequency information is not getting through Ch1 and this points to a failed AD548 but it could also be a part surrounding that chip such as a failed Zener or capacitor. Only high frequency info is getting through and this is capacitively bypassed around the chip.
We know that something is loading the power supply causing a current limit / shutdown mode, but this could also be due to a failed component in the power supply causing it to drive too hard and shutdown.
The two symptoms together could be caused by one part failed in each section mentioned or two separately failed parts in the two separate sections (more unlikely).
If I understand correctly, the capacitor discharge went through the probe ground and somehow completed a circuit back to the capacitor. This surge would have propagated through the BNC ground connector back to earth ground so we can assume that the other end of the capacitor was somehow grounded. It would also have induced a voltage into the center conductor of the probe wire, possibly taking out the buffer chip.
IC's can fail several ways, they can have an input transistor fail from the transient and still draw the correct current or they can destroy the part causing it to present a hard short or a soft short to the power supply. If a hard short, the chip will be colder than normal but some other part will absorb the current and get hot, if a soft short,the IC will be hotter than normal. It can also have a normal temperature and be failed.
Somehow Ch1 must be isolated from the power supply to see if it is causing the load condition.
Alternately you can test the voltage +12AT1 and -12AT1 that supply power to the IC, pins 7 and 4 w/ respect to ground. These voltages are also available across C1085 and C1087 as shown on the attached schematic.
The fact that the other channels use the same +/- supply voltages and work properly points more to the failed chip as open. But something is still loading the supply. Look for any part that seems to be running hot compared to other parts in the other channels and the unit in general.
Testing the power supply can be difficult. Read the attached to see if anything is worth doing as it is a lengthy procedure.
Cheers ION
Dear ION.
Many thanks for the on going info.
My poor old eyes are having a terrible time!! I got +11.9 on C 1085 and -12.00 on C 1087. Is that 0.1 V a clue??
As stated previously I have been feeling all over the board with no sign at all of any noticeable excess heat. One thing that is noticeable is that the scope is only staying on for shorter periods than earlier!!
Cheers Grum.
Quote from: Grumage on 2013.11.29, 18:30:13
Dear ION.
Many thanks for the on going info.
My poor old eyes are having a terrible time!! I got +11.9 on C 1085 and -12.00 on C 1087. Is that 0.1 V a clue??
As stated previously I have been feeling all over the board with no sign at all of any noticeable excess heat. One thing that is noticeable is that the scope is only staying on for shorter periods than earlier!!
Cheers Grum.
Those values are within spec...but do they stay at that value even when the supply goes into squeal mode?
As I mentioned earlier, troubleshooting the power supply is difficult. It can be tested separated from the unit with appropriate dummy load resistors, but this is rather difficult. Also a procedure to check the current drain on each individual supply voltage is noted at the bottom of the pdf....also a difficult enterprise.
Fluke also sells an extender board to facilitate testing the supply, but this still (I presume) allows normal circuit loading, probably just makes it easier to probe. You could probably get some info from Flukecustomer service as I am sure they have seen this before.
Yes, squinting at all those tiny surface mount parts can be tough on old eyes (like mine also, 67)
I
do not recommend the shotgun approach of random part replacement in the power supply, as the problem might not even be there and you will wind up with a damaging of a the pc board, and probably still not solve the problem.
I'm stuck for ideas right now but hope to be unstuck in a bit. Maybe someone else watching this has had similar problem and can chime in.
More to come
Cheers, ION
Edit: I have been referring to this as a TEK skope, should be Fluke / Phillips
ION
can ch1 power supply be removed without affecting the other channels? if so then we will know if it is the supply or just ch1 !!! just my thoughts to narrow it down to which board.
regards
Mike
Dear ION.
Message received and understood.
Voltage drops to +/- (SAME) to 0.95 in squeal mode!!
Also read Mike's post, thanks BTW. There are two Ribbon flex onto the board one small and one large. I wondered if they could be disconnected??
Cheers Grum.
Quote from: Grumage on 2013.11.29, 19:01:51
Dear ION.
Message received and understood.
Voltage drops to +/- (SAME) to 0.95 in squeal mode!!
Also read Mike's post, thanks BTW. There are two Ribbon flex onto the board one small and one large. I wondered if they could be disconnected??
Cheers Grum.
So it seems in squeal mode the power supply collapses. +/- 12 volts feeds all input channels so only one supply for all thus hard to disable. (to answer Mike's question)
Now the challenge will be to keep the power supply up (find the ps fault) and troubleshoot ch1 later as it seems to be a separate problem.
Need to find out what the cables are for before randomly disconnecting.
taking a break!
OK looking at the diagram briefly i would cold check the 2 resistors in the primary drive R1007 & R1008 both are 464K and will be prone to go high resistance as they warm and maybe high anyway even when cold, be carefull in case there's charge in the mains smoother caps.
In fact measure the voltage across each resistor to check the voltage first you dont want to damage the meter on ohms range during resistance measuring O0
Quote from: Peterae on 2013.11.29, 20:18:46
OK looking at the diagram briefly i would cold check the 2 resistors in the primary drive R1007 & R1008 both are 464K and will be prone to go high resistance as they warm and maybe high anyway even when cold, be carefull in case there's charge in the mains smoother caps.
In fact measure the voltage across each resistor to check the voltage first you dont want to damage the meter on ohms range during resistance measuring O0
R1007 and R1008 are startup resistors that provide a trickle of current to kick the oscillator into operation by biasing V1013 up , after which the positive pulse on top of V1009 via R1002 and C1012 keep it going .
How do you reason that it would be these resistors, when the supply starts normally and then goes into an alternate mode of oscillation.
I would suspect those resistors if the unit failed normally and was dead and would not start up due to many hours of operation causing upward drift of those values, in which case the oscillator would not start at all. I have seen this type of failure.
But the oscillator starts normally then goes into a different mode after about two minutes?
I suppose it is worth checking, those resistors probably will read higher than 464k due to age and heating and will probably need to be removed to get a proper reading.
This power supply oscillator has a bit of complexity and interactive functionality despite it's apparent simplicity, especially the functions of V1007 and V1014.
Does it hold to reason that a surge in the BNC ground path back to the mains would take out those resistors or cause them to change value?
Just curious, Peter, as you have probably seen a lot of this type oscillator if you worked in repair, and may have more experience with it than I.
The circuit is quiet a bit different to what i am used to working with, but we have had strange things with the start resistors and almost as a matter of course we would check them anyway even if not related just in case they were causing strange faults, what you need to remember is that it could take me day's to find a hard fault with a diagram, but i could fix 20-30 TVs a day easily by following instink and past experience. I would personally change C1012 or at least have a good sniff(very fishy smell if leaking) and look at it's base for a brown patch.
Sorry i know you got this one covered i should not have jumped in.
Quote from: Peterae on 2013.11.29, 22:41:25
The circuit is quiet a bit different to what i am used to working with, but we have had strange things with the start resistors and almost as a matter of course we would check them anyway even if not related just in case they were causing strange faults, what you need to remember is that it could take me day's to find a hard fault with a diagram, but i could fix 20-30 TVs a day easily by following instink and past experience. I would personally change C1012 or at least have a good sniff(very fishy smell if leaking) and look at it's base for a brown patch.
Sorry i know you got this one covered i should not have jumped in.
Not at all "covered" as it is not repaired yet, and your experience is well respected, you are welcome to jump in since I have no suggestions outside of what I have outlined.
Yes it is true that instincts can do the job faster than pouring over each part in a schematic. In this case it is necessary to retrace the exact path of the current and take a good guess as to which part may have been most likely affected.
This kind of fault, the unusual discharge of a highly charged capacitor into the ground creates a current spike that can propagate anywhere and destroy a most unlikely part. That's why I asked Grum to try and retrace the exact path of the surge through the probe ground, BNC, circuit ground and back around the loop such that we could take a guess as to the most likely to be damaged part.
This is not at all like a normal age related failure mode, and as such it is very difficult to find.
The nose is definitely a help here, the smell of a hot part can lead to the solution, also an IR thermometer helps, but you need to know if the part is being heated beyond it's rating.
Please do comment Peter as we can use your help, instincts, and experience.
I just spent a few minutes trying to to catch up on this subject.
Going by my understanding of the symptoms, I would look into the PS voltage control feedback circuitry.
This suggestion is due to the thought that a capacitive discharge through the shield/ground-plane is the culprit. I would suspect the PS feedback opto (H1046) and/or almost everything connected to it.
A repeating squeal from an non-isolated PSU or practically any oscillation of rail voltage would make me go after voltage control and anything with designed in isolation between circuits.
This may be completely off track because it would all depend upon the part layout and ch1 shield connection relative to the rest of the ground-plane and path to real Earth ground.
yes indeed WW i would definitely be checking the voltage across R1049 to see if the current is creeping up across the opto LED especially as the other half of the 358 op amp is used for line triggering and is connected directly across the mains via caps.
I think it is logical to believe something is temperature sensitive due to the varying times it takes to shutdown, i would get the machine cold and see if it takes longer to shutdown (freezer spray the 358 and opto) and hair dryer the 358 & opto.
From WW:
QuoteThis may be completely off track because it would all depend upon the part layout and ch1 shield connection relative to the rest of the ground-plane and path to real Earth ground.
Agreed it is important to identify the complete current surge path.
Notice the positive 12 volts is used for reference and feedback via the opto isolator. This keeps all the secondary windings nominally regulated. Since the +12 and -12 are regulating to within 100 millivolts on startup and for two minutes thereafter, I kinda ruled this out, but it is always a possibility.
Usually if the feedback fails voltages will soar a bit until shutdown, this would be a good clue if it could be noted.
The cold test is a good idea. If it is really cold outdoors, let the scope sit outside for a bit then see if the run time extends.
Cold by outdoors or spray is a good idea, but should be used carefully since the sprayed area will acquire condensation which could then cause tracking and migration of metals in DC circuits.
My best guess at this time, assuming there is not a part shorting beyond the power supply:
If the bias FET V1016 were leaky, it would begin over biasing the drive transistor V1019 preventing full cut off, heating it and causing core saturation, squealing and a sudden drop in output. Scoping the secondary winding waveform would be the best way to identify this, as scoping the primary is not recommended without isolation transformers. A wafer puncture wound of the FET could cause this. Puncture wounds grow more leaky with time and temperature. The excesss current would keep the SCS biased close to turn on at the edge of what should have been a linear ramp across the transformer current sense resistors, but is now just a pedestal close to SCS firing threshold.
The current sense resistors R1023, R1024 which should show a linear ramp, then cutoff of current when the SCS V1014 fires at the prescribed current of around 2.5 amps, are now seeing a slowly rising bias as the steady state conduction builds in the output transistor V1019. When this current builds high enough, the SCS V1014 is not reset or is only reset at the very edge of the remaining ramp, which now rides on top of the current pedestal.As I said priorly, I have designed these type of circuits using an SCR turnoff (instead of SCS) and a resistor bias in place of the FET. The use of an FET in the location shown is definitely the "Achilles Heel" of this circuit design. They are using an SCS so that they can surely reset for the next cycle, which is often difficult with just an SCR once the holding current is exceeded.
Question to Grumage: does the squealing continue or does it stop completely after the two minute warmup period?
If the PS unit could be removed and fired up with dummy loads as outlined in the manual, that would definitely isolate the problem, but that is a real pain to do.
Normally, it isn't difficult to temporarily replace the power supply with feeds from a bench supply. This is a great way to isolate problems by removing whole blocks of the block diagram.
Worst case, you'll have to lift a couple of leads and connect your bench supply at those points. One advantage is you can then see exact current draw and limit it to prevent further damage.
Don't discount the opto as these and most types of op-amps (especially 358, 741, 07x, 08x, etc.) and electrolytics are the first things we replace when a device hits one of our service benches.
The electrolytics die with simple age (their shelf-life is actually shorter than their in-use life). Anything with GaAs and many MOSFet junctions of all types suffer from simple age problems (normally they last a long time but their susceptibility to failure from transients increases drastically over time).
I've been fighting my garage door opener since the lightning strike several months ago. The signal sensitivity dropped and continued to get worse (remote controls). Yesterday I PM'd the damned thing by replacing all of the op-amps, electrolytics and signal transistors.
The electronics work better now than they ever did - unfortunately, the gearbox cracked during my range testing :'(
WW
A good 5 minute epoxy will make your day for the Gear box repair,a quick solvent clean up on the outside get a bit of fiberglass matt to hold the epoxy and add some extra strength.
whallah.......
Chet
Great suggestion, ramset.
I wish the same would work for the missing teeth on the nylon gear set :(
--------------
I just checked our repair line-card for Grummage's scope. It was black-listed several years ago because of the non-isolated supply and other design issues. Normally, I can grab a list of common failures with corrections and suggested parts. This is what happens when your employer touts reliability of repairs and only sells to industrial and government customers. :-[
Dear All.
Although on line I have been underneath my Son's car most of the day. I have just got back to my desk.
I really DO appreciate your help. Now to answer your question ION, There is no squeal from the PS until the component?? fails. It is then that the squeal appears and the voltage drops off.
I am such a novice with thing's electronic and I am REALLY daunted about continuing!!
Dear WW.
As regards your Nylon gears teeth is there any "meat" underneath where the teeth were?? I could suggest drilling, tapping and screwing Nylon studs into the holes then filing the tooth profile. This has been something I have done on more than one occasion when a gear has stripped and a replacement was not forthcoming!!
Cheers Grum.
Quote from: Grumage on 2013.11.30, 19:41:48
There is no squeal from the PS until the component?? fails. It is then that the squeal appears and the voltage drops off.
Im still leaning toward a bad cap. More than often, a semiconductor issue tends to be an issue 100% or the time. But a cap that has been damaged somewhat, will show itself once heated up. In these cases the temp difference may not be easily noticeable but enough to induce a short in the cap once heated.
I could be wrong 'here'. Say for instance, you turn the unit on, then a time period goes by and then squeal. Then turn it off and then back on right away, is there still a time period before squeal? I say because the cap should have already heated up and the no squeal time period should be shorter after re turn on, being that the cap already had a chance to heat up. ;)
But if the time period is the same after on, off, then on again, then I might say that it happens after a portion of circuitry is initialized, possibly a normal function of the circuitry.
Also you mentioned shielding covering some circuitry and is soldered down? First thing I would do is unsolder and lift the shield and look for damage on the board, if you havnt seen any on other readily visible parts already. The amount of energy from the cap could have blown traces on the board, possibly even shorting to the soldered shield. Even then, there may be bad components.
mags
Dear Mags.
Thanks for the reply. Strangely there is a longer period of time before the next squeal occurs !! I have also removed the soldered cover ( it was where ION said the little chip was ) and no sign of damage there either !!
I am truly lost at the moment !!
Cheers Grum.
Quote from: Grumage on 2013.11.30, 20:29:46
Dear Mags.
Thanks for the reply. Strangely there is a longer period of time before the next squeal occurs !! I have also removed the soldered cover ( it was where ION said the little chip was ) and no sign of damage there either !!
I am truly lost at the moment !!
Cheers Grum.
Maybe best to contact the company to see what they will do for you in way of repair.
I have a Novation synth keyboard than went out on me. The 5v supply was being dragged down. My Wavetek meter measures down to .01ohm, 10 years old, so meters today should be able to do it. But it can help track down shorted parts. I tracked the short to a tiny surface mount cap next to an SM processor. Once the cap was removed, the short was still there, it was the IC.
Cant get it. I called Novation and they referred me to a local repair station. $350 unit fixed for less than $100.
Mags
Yes, leakage building in electrolytic caps is common after warmup and can be excessive if they are failing.
The section in the manual regarding the power supply outlines a method of removing each smoothing inductor and measuring the current in each section of the isolated secondary of the power supply, Page 8-9. This is messy and can destroy the board if not done very carefully. I don't recommend it. Better to just remove the board and see if the supply runs without going into squeal mode. This should vindicate all components on the PS board, so the progressive load must be on the main board. Running the supply removed and with dummy loads is a better method.
Logic dictates that the malfunction must be isolated to either:
1) a faulty component (temperature dependent) on the mains side of the converter, primary side of T1001
2) a faulty component on the isolated secondary of T1001, causing one of the supplies progressively drawing more current.
At lighter loads, the operating frequency will go up. Typically it is between 20 to 50 kHz.
A way to measure the rectifier diode temperatures of each supply rectifier would also give some clues.
Due to the capacitors from earth ground to either side of the line, the transient could have easily ripped into the power supply degrading the FET or drive transistor as stated in my earlier post.
Grum: even though there was no visible damage to the AD548, it could be failing, after all Ch1 low frequency buffer is not working.
Dear ION.
So, and please pardon me for being "Simple" !! can I just slide out the PS board from the large ribbon socket and power up just the PS and HT circuit ( As the HT circuit is on the same board ) ?? Would an open circuit be ok??
My own thought is that if the fault is on the PS it will just start to squeal as before??
Cheers Grum.
Quote from: Grumage on 2013.11.30, 23:12:39
Dear ION.
So, and please pardon me for being "Simple" !! can I just slide out the PS board from the large ribbon socket and power up just the PS and HT circuit ( As the HT circuit is on the same board ) ?? Would an open circuit be ok??
My own thought is that if the fault is on the PS it will just start to squeal as before??
Cheers Grum.
Yes, this is correct logic, however it is sometimes dangerous to run a supply with no load. Also the CRT HV section may go too high and break over since it is meant to operate with a load.
Add to this that you will need a ground to the power supply if you keep the HV and filaments connected to the crt.
This is why loads are recommended and specified in the manual. No easy ways here. Try it at your own risk but it is not my recommendation.
Grum
These people will give you a free repair quote if you can get it to them, but how far is Wigan from your location.
http://www.tercalibration.com/ContactUs.aspx
Quote from: Peterae on 2013.12.01, 09:10:59
Grum
These people will give you a free repair quote if you can get it to them, but how far is Wigan from your location.
http://www.tercalibration.com/ContactUs.aspx
Good find Peter
I would say about 1hr from Grum
regards
Mike
They seem like a fair company, they even calibrate while you are there, maybe they would look at it while there, they even make tea ;D
First though you need to get it back together without making it too obvious it's been apart, there's nothing worse than the sinking feeling you can get when you take the top off and see cowboy hoofs all over the pcb ;D no disrespect meant LOL you would be surprised to know that it takes a real pro to put all the screws back in place.
If you can get your hands on another scope, you can make a sniffer loop to determine if the oscillator is totally quitting after two minutes. It seems to do this from previous info (squeal quits, +/-12 volts fully collapses, scope totally dies, no display)
When and if you determine the oscillator is totally quitting then:
measure the voltage on the current sense resistors R1023, R1024 using an ordinary DMM when it has shut down. Exercise care, this part of the circuit is at mains potentialIf there is a steady DC voltage on those resistors after the oscillator has shut down, the problem is as I have outlined in bold in the post # 30 and we can do some calculations to determine the saturation current.
quote:
QuoteMy best guess at this time, assuming there is not a part shorting beyond the power supply:
If the bias FET V1016 were leaky, it would begin over biasing the drive transistor V1019 preventing full cut off, heating it and causing core saturation, squealing and a sudden drop in output. Scoping the secondary winding waveform would be the best way to identify this, as scoping the primary is not recommended without isolation transformers. A wafer puncture wound of the FET could cause this. Puncture wounds grow more leaky with time and temperature. The excesss current would keep the SCS biased close to turn on at the edge of what should have been a linear ramp across the transformer current sense resistors, but is now just a pedestal close to SCS firing threshold.
The current sense resistors R1023, R1024 which should show a linear ramp, then cutoff of current when the SCS V1014 fires at the prescribed current of around 2.5 amps, are now seeing a slowly rising bias as the steady state conduction builds in the output transistor V1019. When this current builds high enough, the SCS V1014 is not reset or is only reset at the very edge of the remaining ramp, which now rides on top of the current pedestal.
While you don't need to know how it works to repair it with a shotgun approach, it is really quite necessary to fully understand the operation of this switcher circuit to make a
proper diagnosis rather than guesswork. It's operation is well explained in the manual, and it is a good education to learn this for future reference, since you do build power switchmode circuits, besides it is a rather elegant design worthy of understanding.
If you have a IR thermometer and can (non-contact) measure the temperature of the drive transistor and FET you will probably see an excess temperature rise in either of these parts which will stay hot after the scope has quit.
Dear ION.
"It seems to do this from previous info (squeal quits, +/-12 volts fully collapses, scope totally dies, no display)"
Sorry it is the other way round. Scope comes on with various relays clicking then settles with no noise then there is a relay click followed by the high pitch squeal and the trace collapses.
https://www.youtube.com/watch?v=BFZrJpxbnl4
It would now appear that things are getting worse as the scope falls into squeal mode almost immediately!!
Cheers Grum.
Quote from: Grumage on 2013.12.01, 16:51:15
Dear ION.
"It seems to do this from previous info (squeal quits, +/-12 volts fully collapses, scope totally dies, no display)"
Sorry it is the other way round. Scope comes on with various relays clicking then settles with no noise then there is a relay click followed by the high pitch squeal and the trace collapses.
https://www.youtube.com/watch?v=BFZrJpxbnl4
It would now appear that things are getting worse as the scope falls into squeal mode almost immediately!!
Cheers Grum.
Are you saying the squeal starts exactly when a certain relay clicks, and then some time? later the scope goes dead but is not present before that? The exact sequence of events starting at power on is very important. An exact timeline of events would be helpful, and identification of which relay causes this.
Dear ION.
Button ON... graticule illumination...... and voltage range selection relays click. And then almost immediately after..... all go off and squeal starts.
As stated earlier the time period has now become a lot shorter.
Cheers Grum.
PS. The relay is not, IMO a cause but rather a consequence of the PS shutting down!!
Quote from: Grumage on 2013.12.01, 18:21:38
Dear ION.
Button ON... graticule illumination...... and voltage range selection relays click. And then almost immediately after..... all go off and squeal starts.
As stated earlier the time period has now become a lot shorter.
Cheers Grum.
PS. The relay is not, IMO a cause but rather a consequence of the PS shutting down!!
Does squeal continue on? for how long?
Quote from: ION on 2013.12.01, 18:41:31
Does squeal continue on? for how long?
Dear ION.
Yes the squeal continues until I press the power off button.
Cheers Grum.
Quote from: Grumage on 2013.12.01, 18:59:15
Dear ION.
Yes the squeal continues until I press the power off button.
Cheers Grum.
This tells me either:
1) the scenario I outlined earlier is correct or
2) the squeal occurs because the optoisolator is not pre-emptively shortening the ramp cycle because one of the secondary supplies is not available. If the secondary voltage collapses for some reason (shorting part) the opto isolator will turn off, allowing the ramp to run to the end until SCS trips from overcurrent in the CSR resistors.
The function of the optoisolator H1045 is to shorten the ramp by turning on the SCSS before it reaches fault trip current on each cycle, thus regulating the supply since it is driven by sensing two of the voltages, the +10 and the -12 as reference. If it is not pre-emptively shortened, the inductor has not enough time to dump its remanance magnetism, therefore it saturates causing the squeal.
This does not necessarily mean the opto is bad, it could be any of the parts in the feedback loop or non of the feedback parts since you stated earlier that the squeal starts when the secondary +/- 12 collapses. The feedback loop may actually be trying to do its job by allowing extension of ramp time until it senses the target voltage is acquired. In this case, target votage is never achieved so it runs on to the end of the ramp, then trips out on over current, then starts another cycle.
Understanding how this part of the circuit works is crucial and will lead you to the goal.
Check the voltage on C1012, it should be around +15 volts per the schematic or the ramp shortening via the opto will not occur.
Then test all of the power supply secondary voltages to see which of those collapse completely and which try to survive.
Dear ION.
Thank you for taking the time, it is really appreciated.
It is late now so will take a look at the points you have raised tomorrow.
Again thanks, Cheers Grum.