#32 - Fluke 341A voltage calibrator repair

hello we have another piece of vintage test equipment for repair this time it is fluke free for one a DC voltage collaborator it is a precision six digit voltage source with free ranges ten hundred and thousand volts I picked it up on eBay it was described as used powering up and producing some output apparently they thought so because the meter on the front panel showed some voltage the price was reasonable so I bought it and unfortunately it was slightly damaged in transit because of combination of two things on the one hand the seller was not very careful packaging the thing it was loosely put in a box with some packaging material around it but it was not attached properly with the tape to the unit so the unit was able to slide around in the box and on the other hand the carrier was not very careful with the box so as the result this handle on the right side broke off as well as this chilton bail on the bottom here are the plastic pieces which held the chilton bail and one of them completely disintegrated as you can probably see how I already removed the screws holding the top cover because I wanted to have a look at the extent of the damage before I make a decision whether to keep the unit or not and I also wanted to find the bolt which snapped on this side as you can see on one side of the handle have small piece of it broke off and it is too attached on the other side though the bolt snapped and it fell inside of the unit so I wanted to find it to make sure it didn't short anything so I decided to keep the unit and I negotiated some discount because of this damage I think I can repair these mechanical problems I will need to straighten out the front panel a little bit and I think I can fix this handle as you can see the holes are deep and there is enough thread I think to have longer bolts I can glue back this small piece which broke off just for the looks and then rely on longer bolts to hold the handle in place but let's worry about mechanical parts after we repair the electronics here is one of the pictures of the unit from the auction and as you can see in the handle was intact the Tilton bail was in place and it was set to 0 volts but the meter showed slightly more than 10 volts output but the links between the output terminals and since terminals were not installed so I'm not sure if they tested the thing properly let's find out so here I installed the links between the output and sense terminals and powered the thing up it is still set to 0 volts but look at the meter it goes up and down something is wrong with this unit and look at this it goes out of scale something clicks inside and it goes back I played with it for a while and suddenly it started working well almost you see there is some offset around point eight point nine MIDI volt and it's jumping around quite a bit in this ten volt range the offset means eight or nine counts here in this decimal place and there is no way it could be this far off if it worked properly at most it could be one or two counts off in the last decimal place if we forget about the offset for a moment look one volt two three four five six no problem here hundreds of millivolts no problem here as well tens of million volts single millivolt problem here as well and even here we see that steps to buy 100 microvolts no problem at all and this should be down to the noise we could hardly see if it's right or not and there is another minor problem we are in the 10 volt range now and the decimal point is here but if I switch 200 volts the additional point should move here and this indicator doesn't line up and in the thousand volt range we should see this one and it also doesn't work but these five indicators are lamps not LEDs because this thing is so old so high just these two lamps are burnt and probably this one as well because when I try to limit the current it doesn't light up at all so I think I need to replace these lamps with LEDs and let's try the other ranges hundred volt range so the decimal point should be here so this is 10 volts 20 30 40 50 and so on looks like no problem at all hundred volts and this is false in volt range the decimal point is here so this should be 100 volts 200 and so on no problem at all I have this service manual with full schematics and description of the principles of operation and such and look at this John Luke June 1969 this company has been around for a while look at these specs here in point zero zero five percent accuracy on all the ranges not bad at all for 1969 and for my hundred bucks if I managed to repair the thing and by the way the higher model which is free 43a has seven digits setting and accuracy is 20 ppm on all ranges not bad at all here we have the performance test section in the service manual the first test here is line regulation we are supposed to use an autotransformer to vary the line voltage from 115 volts to 126 and a half volts and we should verify that the output stays within 30 micro volts I can have another transformer and this test doesn't make much sense anyway at this point because as you have seen the output voltage is jumping around in the order of hundreds of micro volts the next test which is load regulation and doesn't make sense for the same reason but the next one after that is ripple and it makes perfect sense we should verify that for this model the output ripple is below 100 micro volts RMS or 1 milli volt peak-to-peak and we should do that with positive output connected to chasis ground also using some sort of pre amplifier is suggested here here is the list of recommended equipment and this pre amplifier is not specified at all except the gain of 1000 and frequencies five Hertz 200 kilohertz I don't have anything like this handy I imagine it should be a precision pre amplifier with low noise low distortions and precision gain let's see what we can do without it so I moved the grounding link to the positive terminal here we are supposed to set 10 volts and I switched the meter into volt AC we are reading around 30 micro volts RMS which is well within specs so the output ripple doesn't seem to be a problem and measuring picture pic with the scope at least with my scope seems to be hopeless look at this with nothing connected and they said to 1 millivolt being a per division and AC coupling we are reading about 1 milli volt peak-to-peak and about 500 micro volts RMS we can try limiting the bandwidth to 20 megahertz this should help a little and now we are reading about 480 oh well about 500 micro volts peak-to-peak and slightly more than 400 micro volts RMS this is a bit better but still not enough this is why preamplifier is recommended in the service manner and that preamplifier must have much better newest floor than this this thing started misbehaving again doing negative then going off scale and then going negative again and so on finally let's have a look inside the precision switches are under this cover so let's take it off here we are and these switches are gold-plated and they look perfectly fine but this section here looks a bit rusty and perhaps needs some cleaning you and here is the bottom of the unit this happens to be a single board not to words as it looked from the top a revision age is this a date code here seventy eight oh three and seventy seven thirty eight and this one in the center and seventy eight oh three on this one seventy seven forty this is the oldest one I see so far seventy-five twenty-one let's look what's under this cover removed the bolts and here we are not shielded by the way just a piece of plastic I would imagine a couple of reasons for this one is to keep the dust away from the sensitive part of the circuit and another is to keep this circuit at the same temperature as much as possible you so let's clean these rusty switches let's use this super contact cleaner with lubricant perfect stuff for the job we should apply a small amount right on the contacts and then let's reach them back and forth a few times and the other section as well and now the second switch this switch has only three positions this is the range switch and the second section of it alright should be good enough the next thing to do I think is to check electrolytic capacitors let's grab this LCR meter and start with these big ones the only way to reach the pins of these big caps is from the bottom and look at this arrangement each cap has five pins and these four pins are connected together I'm going to use these multi meter leads instead of the proper leads for this LCR meter because I need sharp ends to penetrate the coating on the board let's go the default is one kilohertz but for larger caps it's better to switch 200 Hertz and the second parameter now that's set to you sir so how this cap is c1 and it should be 80 micro farad's here we are 87 and the SAR is 1.6 and this cap is e2 and it is the same 80 micro farad's 91 and the SAR is 1.3 not bad at all not a problem I checked all of these big caps and they all look fine I also checked these five blue caps in this section and they also look fine for example this one is C 22 and it should be 250 micro farad's we are reading 242 and the SAR is point 3 ohms I checked capacitors in this section and I think I found a few problems first is these four capacitors are the same type 10 microfarads 50 volts and the one I'm measuring now looks fine ten and a half micro farad's fifteen and a half ohms ESR at hundred Hertz but let's check another one 10 microfarads but almost 30 ohms ESR and the one next to it even worse 50 ohms ESR and another one even worse than that nine and a half microfarads eighty-three ohms yes sir so I think they are on the way out the next problem these two caps are 100 microfarads three volts and let's check this one 135 1 ohm yes sir not a problem this second one for micro farad's hundred 70 almost ohms ESR so this one is completely dead the next problem these two caps are 5 micro farad's 25 volts so let's check this one well 5.4 let's say micro farad's 14 and 1/2 ohms ESR at hundred Hertz and another one 5 micro farad's but 32 almost 33 ohms yes sir so clearly it's not as good as it should be and there is one more cup of the same type here probably not visible on camera right now and the sir is not very good as well somewhere in between those two we measured before and there is one more problem here these two caps they are not electrolytic caps I checked them anyway and I'm glad I did they are point zero 15 micro farad's or 15 nanofarads and let's check this one 15 nanofarads I am at 1 kilohertz now and let's check the other one it is negative why there is negative because if we switch to resistance it's a short and here they are c11 and c10 as you can see in this circuit is completely symmetrical and I checked these diodes and transistors around them they look fine no shorts at all so there is no reason these caps should measure differently at all I went to a local electronics store and picked up some caps and because they had chip I picked up model enough of different types so I can choose and probably it's a good idea to replace all the electrolytic caps here but for now let's just try to replace the ones which are clearly bad and let's see what happens I'd assaulted this c11 cap and sure enough it is shorted so these are point 0 15 micro farad's or 15 nanofarads 100 volts 2 percent and 2 percent means within this range from fourteen point seven to fifteen point three I didn't find two percent caps I picked up three different types these are 5 percent 400 volts and almost all of them are within this range but they are quite inconsistent all over the range these are 10% caps 200 volts but they are very consistent all four are very close to 14.8 there we go fourteen point eight six nine and I don't know tolerance for these guys they are also very consistent the 100 volt caps and all of them are close to fifteen point one fifteen point one seven this one is 15 point one zero so I am thinking about using couple of these or a couple V's and it's obviously better to replace both so they are close to each other and have the same temperature drift and such I don't have unfortunately a full specs for any of those caps instead of these 10 microfarad 25 volt caps I picked up these guys 10 microfarad 63 volt caps and the sir is about 10 ohms this is at hundred Hertz this 100 microfarads 3 volts by sprog we found to be completely dead so I picked up a couple of new Chacon 100 microfarads 25 volts to replace it and the other one which looks okay but I would rather replace both just in case and these are 5 micro farad's 25 volt caps by Spragg which had high yes sir so I picked up 6.8 micro farad's 65 old capacitors also by Spragg to replace them there is no five microfarad value nowadays only four point seven or six point eight and look at this five ohms yes sir and this is at handle it hurts again so I replaced some capacitors here these two point zero fifteen micro farad ones these four 10 microfarad ones these two hundred microfarad ones and these free 5 micro farad capacitors which I replaced with 6 point 8 micro farad ones maybe waita I will replace all the electrolytic s-- but for now let's leave them alone and let's test the unit oh look at this we are at 0 volt setting and before we had about point eight or even point nine milli volts offset here it's completely gone now now the offset is negative one or two micro volts which is perfect here I have this table in the service manner output voltage requirements and there are two columns one is for an instrument that has just been calibrated and column two is tolerances according to specifications so in this 10 volt range at zero setting we should be within plus minus 10 micro volts and we are within plus minus and well up to 3 micro volts which is well within the tolerance even for freshly calibrated instrument let's check hundred and thousand volt ranges at zero settings we should be within plus minus 20 micro volts and plus minus 60 micro volts and it will the range minus 10 micro volts well within the tolerance and thousand volts minus 2 or 3 micro volts perfect let's move on next step is 10 volt range 10 volt setting we should be within plus minus 200 microvolts in the column 1 or plus minus 1 millivolt in the column 2 so 10 volt range 10 volt setting 300 micro volts out which is slightly out of the column 1 but well within column 2 next here is 10 volt range 5 volt setting we should be within plus minus 200 micro volts or 500 micro volts here 200 microvolts out which is right in the edge of the column 1 next is hundred volt range hundred volt setting and we should be within plus minus 2 millivolts in column 1 or 10 millivolts : - lets see hundred volt range hundred volts setting spot-on next is hundred watt range 50 volt setting + - 2 millivolts or 5 millivolts in the colon - let's see 1 millivolt out perfect next hundred watt range 10 volt setting plus minus 500 micro volts calling 1 1 millivolt column - here we are and it micro volts well within : 1 next hundred watt range 5 volt setting plus minus 300 micro volts or 500 micro volts so zero here 5 here we are 500 micro volts out which is right on the edge of the column - next thousand volt range thousand volt setting we should be within 20 millivolts column 1 or 100 millivolts : 2 thousand volt range of the world setting and we are 30 over 20 millivolts out which is right on the edge or slightly out of the column 1 but well within column 2 next thousand world range 500 world setting plus minus 20 millivolts again or 50 millivolts : to 500 and it's perfect next thousand volt range hundred volt setting plus minus 5 millivolts or 10 millivolts and it's perfect in the last one and I believe there's a typo it should be thousand volt range because of the decimal point if it was hundred volt range decimal point should be in different place so thousand watt range 50 volt setting should be plus minus 3 millivolts : 1 or 5 millivolts : 2 so 50 volt setting and we are 5 millivolts out which is right on the edge of : 2 so if we trust my meter this unit is well within specs and unfortunately I don't have a better meter ideally a meter should be much more accurate than the unit we are checking these are comparable in accuracy which is not ideal but that's the best I can do at the moment I took the front panel off to straighten it out and fix the handle I also need to replace these lamps with LEDs all right I think I managed to get it into good enough shape and I also managed to get this piece of broken bolts out all right the handle is in place that broken-off piece i glued with a drop of super glue as I said just for the looks and then slightly longer bolts hold the handle securely in place no problem so these lamps are just sitting in short pieces of flexible tubes and shining into these diffusers connectors are used on the pins so they are really easy to replace and the idea is almost perfect a slightly loose so I might need to put a drop of glue or something to hold that in place and I also need to add current limiting resistors these are 28 Volt lamps I installed five white LEDs instead of the lamps with the decimal points and as you can see this one is shining and the diffuser is yellow i salvaged these guys from some sort of a flashlight they were great here I applied a drop of glue on them to keep them from falling out and I added 3.3 K resistors to limit the current at about 8 milli amps now look at this I borrowed this meter 3 4 4 6 1 a from a friend of mine it is still branded agilent with a keysight sticker on it technically it's about the same accuracy as my meter but it's a new model and look at this certificate of calibration dated 20th of august 2014 so it was calibrated by a keysight slightly more than a year ago and this folks is as good as it gets I warmed up the equipment for more than half an hour and we are at zero sitting 10 volt range and we are reading about negative 5 micro volts which is well within spec for the fluke and look how these meters agree almost perfectly within one micro volt so let's try adding 10 micro volts which should bring us to positive 5 which it did 20 micro volts 50 and we still see this 5 micro volt offset 90 still about 5 micro volts down let's try a hundred micro volts 400-800 still about five mic levels down 1 millivolt 3 millivolts five perfect 8 10 millivolts 50 millivolts 90 hundreds four hundred nine hundred perfect one vault - 5 10 meters agree almost perfectly let's go 200 volt range 10 volts 20 volts 90 volts and in volts let's go to thousand volt range hundred volts three hundred seven hundred thousand so this is 10 volts and if we add 1 millivolt to millivolt this is all perfect and meters agree very well I am very happy with all this I managed to glue back together this foot with superglue it was broken into several pieces and I filled this missing piece of plastic with epoxy and these feet are supposed to sit on the surface like this and this second foot has this part broken off partially so I cut in half a rubber foot like this and add it here and I will cut this one as well and add to this one and I have more of different sizes which I think I will add to the back of the unit instead of the Machine rear feet it had only two feet on the front where the children bail here are the front feet installed and the real ones as well I hope you enjoyed this look at such a fine instrument by John fluke manufacturing company isn't this amazing the unit was built in late 70s so it's about 37 years old but after replacing a few failed capacitors it is completely in spec I didn't have to touch any adjustments and for the most part it met requirements for freshly calibrated instruments unfortunately I only had six and a half digit meters to check the calibration but the meters agreed so closely that I am quite confident that everything is great thank you very much for watching if you liked the video and give it a thumbs up and subscribe to the channel bye

Channel: FeedbackLoop

Views: 12,415

Rating: 4.9596772 out of 5

Keywords: electronics repair, lcr meter, voltage reference

Id: mxz3-fGs1Lk

Length: 45min 12sec (2712 seconds)

Published: Thu Nov 12 2015