LCR vs ESR Meter

[requim]

I'm wondering what the right tool is for the job.  I am just learning how about electronics and at this point my repair work qualifies as following others directions.  I am primarily interested at the moment in being able to repair older/broken items.  I'm of the understanding that capacitors have the highest failure rates in most (all?) electronic circuitry and that in order to detect failure I need to measure both the capacitance as well as the ESR in order to have a clear picture of whether the cap is good or bad. 

I've also read that you can apparently use a function generator and an oscilloscope to get ESR values as well, but I believe that means having to probe the same cap twice, once with the DMM to get the capacitance and again to get the ESR.  So while it would be cheaper to simply do the two probe test, I'd rather not have to measure it twice.  Therefore I'm looking at devices capable of measuring ESR and I've found ESR meters, which mostly appear to be DIY projects along with some pre-assembled versions and LCR meters which are mostly in the realm of major manufacturers.

So my question is, what's the right tool for the job?  I'm under the impression that if I'm doing design and building work that an LCR meter is the right tool for the job, but for repair work, an ESR meter is good enough.  Is my analysis correct?  And if so, what meter do you recommend?

I've been looking at the BK Precision 879B that Dave reviewed along with the Agilent U1732B.  Been hoping the price would drop to clearance at Granger so I can pick one up and make the whole decision making process easier.   I've also looked at the Agilent U1733C and think that if I'm going to spend $300 on the BK, I might as well spend $400 on the Agilent.

As for ESR meters I've seen the Blue, with mostly positive reviews, the Peak Atlas, with mixed reviews, and then I've also seen this one particular one which looks nice on ebay:

http://www.ebay.com/itm/ESR-Meter-Electrolytic-Capacitor-Meter-Green-Black-bg-LCD-DC2-1x5-5-/130591361363?pt=LH_DefaultDomain_0&hash=item1e67da0953#ht_3348wt_1416

They have the same meter with three different lcds.  The price seems reasonable enough, though I'm wondering if they're just building from a kit and selling them because I've looked at some of the schematics posted by Kripton2035 in another thread and it looks very similar.

Thoughts? Suggestions?  Recommendations?  Advice?

[alm]

I agree with your impression that capacitance measurement in mainly useful for design (eg. unmarked parts, or matching). For repair, ESR is a much better indicator of how healthy electrolytics are. Don't have any suggestions on specific equipment, I own an AADE LC (no R) meter mainly intended for RF parts. Good resolution down to 0.01 pF or so, but only 10uF(?) max. I use my DMM for larger electrolytics if I want to measure them (quite rare). For ESR I use a version of the Bob Parker ESR meter.

[amspire]

LCR meters are nice, but if you are starting out, a LCR meter is probably not a high priority.

For repairing devices, a in circuit ESR meter is very handy as it allows you to check all the electrolytic caps, and replace any that are dying. The Blue ESR meter is very good, but something you build yourself for $20 is equally good. Accuracy is not really required for an ESR meter - you could get away with 20% accuracy without a problem. A digital readout with lots of digits is cute, but not worth paying extra money for.

Almost all decent multimeters have a capacitance range, and it is perfectly adequate if you need to check a capacitor's value.

Inductors, by their nature, don't change their value. They either work or have a major failure - open circuit, shorted windings, overheated.  You can do most repair work without an LCR meter for checking inductance.  If you have a signal generator and resistors, you can always rig up an inductor test if you need it.

Don't get me wrong - many of those new LCR meters are totally brilliant, and if you are pining after one, then get one.  It is not an essential repair tool.

Richard

[eternal_noob]

I have to agree with alm and amspire here, and I would definitely go for an ESR meter. The lower resistance it can measure the better it is in my opinion.
The capacitance of almost any electrolytic cap can easily be figured out measuring time constants, and I honestly don't understand why cheap DMMs have a capacitance range at all, since they rarely have any accuracy worth mentioning on the lower ranges.
If you are repairing older stuff, like tube radios and such, brew your own meter and save money.

[requim]

As far as equipment goes I've been on a rampage buying gear.  I bought a few items that are broken or have problems with the hope that by working on them I can build my knowledge and skills while at the same time saving quite a bit of money.  I have most of the gear or items that Dave addressed in his building an electronics lab with one major exception being a DSO.  Quite frankly I don't want to buy a DSO till I have enough experience on an analog scope.  I figure that for my purposes the DSO is just a nicety.  Please correct me if I'm wrong.

Hence for that reason I have been looking at a ESR meter or LCR meter.  An earlier poster mentioned he had a kit LC meter and another kit ESR meter and looking at them to buy the kits and have them shipped I'd spend ~$240 and at that price I'd just as soon buy a BK Precision 879B for another $60 which I believe to be superior to the other two units as well as having a warranty, support, etc built into the price.

Realistically I'm just pining for a nice LCR, however my pocketbook is pleading otherwise. 

@amspire I'm curious where I can get an ESR meter kit for $20 because the cheapest I've found is closer to the $80 mark.  I've read of people building ESR meters from schematics for less.  And in another thread I saw you discussing with Kripton2035 about his project to try to build one with six components, but I haven't heard of him having success with that yet.

As for digital versus analog, I just prefer digital.  I find it faster to read.

Any opinions on the ESR I linked to my original post?  It's less expensive than the Blue and the Peak Atlas and it offers simultaneous readings for capacitance and ESR like the Peak Atlas.

[Zad]

To be honest, when I was doing repair work, I never used an ESR or an LCR meter. You scope a voltage bus, and see ripple or noise, or see the bulged electrolytic and swap it. Sometimes a meter just won't see the non-linearity that happens under load anyway.

I'm afraid people are getting the wrong impression from Dave's test equipment reviews. You lean diagnosis and repair with your brain, not with more and better gear. I would spend 20 bucks on some components and a case and make my own ESR tester, they are about as simple as it gets.

[Psi]

You lean diagnosis and repair with your brain, not with more and better gear.

I totally agree,  repair and fault finding is a mix of understanding how the circuit works and experience/instinct about where to start looking first, given the particular fault.

Good test gear just makes it easier and faster.

[amspire]

If you want to buy a LCR meter, save up for a proper one.  One that provides a two value reading for inductance and capacitance, rather then a single value.  Also one that has proper specifications.

The cheap LCR kits claim accuracy, but tests I have seen show they are often not accurate over the whole range, and with inductors and capacitors that have a relative high lossy component.  When you test capacitors and inductors, it is important to test at a known frequency (as the values of real components are not the same when measured at different frequencies). With the cheap LCR meters, you don't know the test frequency.

The $20 LCR meter I was talking about uses scrap parts as much as possible (like a  transformer from a dud PC power supply), and a cheap moving coil meter from somewhere like ebay or a VU meter from a scrapped piece of audio gear.  Using a meter movement uses much less power then a digital display, and you are really after a "go/no go" reading.  It is not a device you use every day.  You might only pull it out a few times a year.

Multimeters and an oscilloscope you do use a lot.  When you are repairing, you want to be able to 'look' at what the circuit is doing, and the way you look is with multimeters and oscilloscopes.

A DSO is handy for debugging, especially if you are trying to capture what is going on just before the power fuse blows in a faulty device.  Analog scopes are only useful when there is a stable repetitive waveform to look at. It is good having both an analog and digital oscilloscope, but if I could only get one, I would go for the digital oscilloscope. For repairing equipment, then for an analog scope, a really cheap 20Mhz one is adequate, and if you are lucky, you can find one for $50.

[requim]

@amspire - Well from what it sounds like then at this exact moment I'd be better off with a DSO.  I'm currently working on a broken Tektronix 2235.  I assume it has a short in it because it blows the fuse whenever power is applied I guess a DSO might help me in diagnosing it faster.. 

I believe I know the primary source of the problem (the HV Power Transformer) but from what I've read the caps on them have a pretty high failure rate as well.  I thought about going hog wild and just blindly replacing all the caps, but figured I should just test them with my DMM.  Later I read about ESR and how common it is in older caps as well so I figured I should investigate that further which is where I am at the moment.

Am I making ESR out to be a bigger deal than it is?  Am I making mountains out of mole hills? I figured if I had an ESR meter or an LCR meter I would save myself the step, time, and money of blindly replacing parts that may or may not need replacing.

As far as equipment goes I own one working and one broken analog scope, a signal/function generator, three linear power supplies, a dmm and a malfunctioning bench dmm, a heat gun, and a soldering station.  I think I've got most of the basic major gear already excepting the DSO. 

My projects at the moment surround repairing the broken/malfunctioning gear I've purchased.  Once I've done that I figure I'll look into kit projects or something simple to help me understand basic design and assembly elements.

If an ESR or LCR meter isn't next on the list of major equipment to buy, what is? (ignoring the DSO for the moment)

[amspire]

The good thing about having an ESR is you can probably check all the electro's of something you are renovating in about 5 minutes, and you might easily find some capacitors worth replacing.

Without the meter, you would leave the old caps and it is possible that might lead to problems.  If you have one, and you find high ESR capacitors, it may have paid for itself when you look at the debugging time you have probably saved.

Rather then looking for more instruments to buy, why don't you look for an old good quality bit of gear that needs fixing, that you could resell on ebay if you like. Something that is also interesting, but that is build from discrete parts and that has service manuals you can download. This would include a lot of old HP and Fluke gear.

I will just give some examples. 

In the past, I have seen some Fluke 540B's go really cheap and the cheap ones always need a lot of work.

What is it? Something quite different from what you are used to. It is an AC thermal transfer standard.  Basically it means as long as you can measure DC volts to better then 0.01%, you can measure an AC source to 0.01% true RMS accuracy.  That is probably an order of magnitude better then any modern sub $1000 meter can do. Better then most of the over $1000 meters too.

The old ones always have leaking and shorted NiCd batteries that need to be replaced with new NiCds (about $35 on ebay), and that probably means there will be some corrosion on one of the boards to be cleaned. Switches probably need to be checked and cleaned. Also they require a Mercury battery as a stable voltage source.  Mercury batteries have now been banned, but many of the 1.25V IC regulator have a good enough short term stability as a replacement. If the thermal sensor is dead, then you are out of luck, but the 540B comes with a hair trigger protection circuit, so hopefully the sensor is OK.

Here is another suggestion :

http://www.ebay.com/itm/HP-49G-Graphing-Calculator-Defective-/400250937946?pt=Calculators&hash=item5d30d0a25a

These HP49G calculators had a common problem. An chronic leaking electro (lead type - not surface mount) connected to the backup memory stops the calculator from starting. Replace it and the calculator is like new. If the symptoms are that the calculator won't start, and it looks like it hasn't been used that much,  there is a very good chance it is the capacitor.   $30 is probably a bit much for a dead calculator though. It is functionally about the same as HP's current to of the range calculator, so working it is pretty neat.

Things like HP power supplies. If you find a broken one and fix it, they are easy to sell. So if anything is busted other then the mains transformer or the meters, an old power supply can be fixed.

It all comes down to finding something you will find interesting, otherwise it just feels like hard work.

Fixing the 2235 is a good job, but oscilloscopes can be pretty tough to fix if you are learning - pretty dangerous as well if you have to touch the HV circuits. Might be the type of job you leave for later.

Richard.

[requim]

I feel like I've kind of gone in the direction of which you speak which is why I have been working on the HP/Agilent e3610a, which BTW I replaced the two caps on which eliminated the transient voltages.  I am still waiting on parts from Agilent to implement the service note fixes on.  I have the Tek 2235 I mentioned which I purchased intending to fix. 

I also have a Agilent 34401a DMM which I purchased knowing that it won't settle on a value.  I'm planning on posting about it once I get a couple of these projects off my back.  If you have any ideas I'd be more than happy to hear them since your advice was spot on with the 3610a.

And finally I just purchased today an Agilent 6611C DC Power Supply which has as of yet unknown error with it.  I won't receive it for another week or so so until then I don't have a clue what's wrong with it.  The good thing is that it powers up.

And this whole electronics kick I'm going through has been good because an old 3dfx Voodoo5 5500 video card I received about a week ago arrived with one of the caps broken off.  Luckily I had purchased my heat gun and with less than a minutes worth of work I repaired the card and so far it seems to function normally.  The heat gun itself paid for itself or at least saved me a lot of hassle with the seller.

So to this point I've just been purchasing some broken gear hoping to use it for myself.  I figure I can get things for less than half the used going rate or less, and with any luck can fix them with a few dollars in parts and material.  I'm not really interested in the moment at fixing gear to resell it, but I suppose a day may come along where I might consider doing it just as a hobby.  Personally I tend to collect and keep or collect and give away.  I hate the whole selling process so I avoid it by giving stuff away.  Makes me happy and makes the recipient happy.

Anyway, I will look into the two suggestions you had as possible projects.  My primary reason for wanting to buy equipment is because I want to have the right tool for the right job.  For years I've thought about dabbling/jumping into electronics.  I purchased a radio shack analog multimeter meter years ago along with a soldering stick (or whatever those things are called) and needless to say I was totally disappointed and moved onto other interests.  Dave's blog has opened up the world of electronics to me and given me a template on which I can make some good purchases.

But I digress.  Thanks for the advice and if you have any ideas what would cause my Agilent 34401a to not settle on a value when probing or when nothing is attached I'm all eyes and ears.  I'm really hoping its repairable since I paid a pretty penny for it.

Thanks!

[requim]

@amspire - BTW what is the price range for a good project Fluke 540B?  I see them on eBay right now between $200 - $600 which at those prices I wouldn't be interested in acquiring one to fix.  Too big of an investment in a piece of equipment I have no direct need for at the moment.  I would certainly consider it at a much lower price, but I don't know if that's a reasonable expectation.  Thanks again.

[amspire]

No. A few hundred dollars is way too much. They have probably been fixed up already. It is just something interesting to look out for, but there are other interesting old instruments.

Which country are you in?

[amspire]

If you are in the US, this HP 8012B pulse generator could be a good repair project.

http://www.ebay.com/itm/Agilent-HP-8012B-Pulse-Generator-Parts-Repair-33717-282-/310354922631?pt=LH_DefaultDomain_0&hash=item4842984c87#shId

Manual:

http://bama.edebris.com/download/hp/8012b/HP8012B.pdf

Notes on repairing:

http://www.orionsword.net/Electronics/TestEquipment/HP8012B/HP8012B.html

[requim]

The USA. 

eBay seems to be my primary source to find stuff, though I went to an auction saturday and realized how much cheaper I could get stuff for.  Unfortunately this particular auction didn't have electronics gear but it was something I'm keeping in mind.  In addition I found that the local university has a surplus store, but apparently governments, university departments, etc get first dibs.  I was going to pickup a tek scope but it sold before I got there. C'est la vie. 

I've also sourced some stuff from craigslist from daring participants who were willing to ship items to me, but those are hard to find and with craigslist I haven't been buying broken units since you don't really find that sort of stuff on there.

Another upcoming project is fixing a KitchenAid stand mixer that broken after a few uses, but unfortunately we didn't open it up until well after the warranty had expired.  I don't want to tackle that though till I have a better grasp of what I'm doing and being safe around high live voltages. No use hurting myself unnecessarily if I don't have to.

[requim]

Yeah i was looking at that but then I checked the prices that they're selling for in working condition and the price is only ten or twenty bucks higher.  So far I've been primarily chosing projects where the cost is no more than 1/2 the low used prices for the same item go for.

I will say this after looking at one of the ESR schematics and more importantly the source code I was tempted to look into the project further, because I'm a software developer by profession and I have an interest in actually doing some low-level work.  The idea of customizing the esr meter was intriguing to me and wasn't something I had previously considered with these projects.

Is a pulse generator worth having?  I already have a signal/waveform generator...

[kripton2035]

you can go to my web site and get open schematics for esr meter and lcr meters
grab your own !
http://kripton2035.free.fr/esr-repository.html
http://kripton2035.free.fr/lcr-repository.html

[alm]

To be honest, when I was doing repair work, I never used an ESR or an LCR meter. You scope a voltage bus, and see ripple or noise, or see the bulged electrolytic and swap it. Sometimes a meter just won't see the non-linearity that happens under load anyway.

What about the small electrolytics often found in the feedback circuit of switchers? They often prevent the power supply from starting. ESR is not a complete replacement for ripple measurement, but ripple measurement is not a replacement for ESR measurement either.

I'm afraid people are getting the wrong impression from Dave's test equipment reviews. You lean diagnosis and repair with your brain, not with more and better gear.

No argument here.

[amspire]

Ok. I think I have found something interesting, and it definitely could do with some restoration.

http://www.ebay.com/itm/Lambda-LS-512-Precision-Power-Source-Not-Working-Well-/370513471532?pt=LH_DefaultDomain_0&hash=item564452e82c

It may need calibration, but maybe not - the reading shown on the Fluke multimeter and the setting on the voltage source agree within the Flukes specs.

Also if I was in the US, I would be extremely tempted by this:

http://www.ebay.com/itm/C83940-Lot-2-HP-3455A-Digital-Voltmeter-parts-repair-/390359680872?pt=LH_DefaultDomain_0&hash=item5ae3400768#shId

The strong possibility of getting one HP 3455 voltmeter working - perhaps two would be great. No problem getting the service manual free and it is a great manual. There will also be an HP journal article on this meter that can be downloaded.

This is a really interesting multimeter because it self calibrates from a removeable module that contains the voltage reference and a precision 10 x divider and a resistance reference. If these three things in the reference module are calibrated, then so is the multimeter (except perhaps for the AC converter - I forget the story there).  Now there are fairly easy ways for a hobbyist to calibrate a 10X divider within 1 part in 10^-5 or better, and there are some excellent cheap voltage references that can be purchased on the Internet with sufficient accuracy to calibrate the voltage reference. So this meter can be fully calibrated by a hobbyist, except maybe for the resistive reference that is probably very stable anyway.

Find one of the Fluke 540B's cheap (that I mentioned before) and you can even calibrate the RMS AC to spec all the way up to 1Mhz.

The ratio calibration between ranges fully self calibrates off the 10x divider right from the 100mV range to the 1000V range. You don't need 1000V to calibrate the 1000V range.

Just after this meter, HP went away from this concept of self calibrating ranges to the precision thick film tempco tracking dividers that everyone uses now, so this meter was a first and last. But because it self calibrates, it is great if you don't have a friendly standards lab to calibrate the meter.


A very good meter.

Richard

[Conrad Hoffman]

The 3455 is a great meter- I keep a couple around. The idea of calibrating the module alone sounds great, but nobody does. Just make the adjustments to the module trimpots on the rear using the whole meter against whatever references and resistance standards you have. It's not terribly difficult to service, but definitely not for the neophyte.  I've never found schematics for the reference module that match the actual hardware. You might have to replace the trimpot(s) on the module because they tend to get worn in one spot- the spot where you need them most.

If I had any complaint it's that getting the meter to read the same for both polarities can be difficult or impossible. It's only a couple counts error, but it's annoying to the perfectionist. The meter is big. You better have a deep bench! I had my bench custom built so I could use the old boat anchor HP stuff and still have room to work in front of it.

See my website for how to build a Hamon divider to get /10 and /100 ratios to extreme precision. For resistance standards you just have to go with the best you can get, maybe have a local cal lab measure a couple wire wounds for you.

On LCR meters, did you see the Tenma sold by MCM electronics? It's a whopping $34 for the capacitance-only meter, and not that much more for LCR. IMO, the c-only meter is the better one because it has a wider range.

[amspire]

Conrad,

Just had a look at your website and some really good stuff there.

The Hamon divider page is at http://conradhoffman.com/HamonResistor.html

Do you actually have a GR 1615A capacitance bridge? Pretty amazing devices. You never know when you will need a 10 atto Farad capacitor.

I see the big brother the GR 1621  bridge is still available from IET labs at some unimaginable cost.  That is for when you need a 100 zepto Farad capacitor (10^-19 F )

You are right about the size of the HP 3455A. I used to use them back in the early 80s, and it didn't seem so big then - all the other lab instruments were even bigger!

Richard.

[grenert]

Back to the ESR meter (and ignoring for the moment whether or not the OP really needs one at this point), if you want a nice kit, the EVB ESR meter would fit the bill:
http://clientes.netvisao.pt/greenpal/evb1.htm
49 Euros shipped.

It's the same electronics as the Blue meter, based on the old Dick Smith project.  It differs a bit from the Blue in that it has banana jacks on the meter, for you to plug in test leads, rather than built-in leads.  I like that the Dick Smith design also works as a low ohms meter (down to 10 milliohm).  Personally, I find an ESR meter useful when building small power supplies for projects.  Some low dropout regulators don't like low ESR caps on the output, for example.  In unregulated supplies, I like to use the lowest ESR caps (better ripple filtering), so I can compare the ones in my junkbox with the meter.

Kripton2035:  Nice sites!!!  Thank you for compiling those.
I've spent a lot of time on Conrad's site before, and it is packed with interesting stuff.

[Conrad Hoffman]

Yep, do have a 1615! There's a page on capacitors and LCR meters I've been working on, but it's not ready for prime time. There are no public links to it yet, but I'd be overjoyed if people here would give it a read and tell me what they think. I'll try to incorporate whatever changes make sense. The hidden page is http://www.conradhoffman.com/capchecktut.htm

Thanks,
Conrad

[requim]

@grenert - I looked at the ESR meter you recommend, but the 44 euros they charge to ship that to the US kills any sort of deal.  I imagine that thing probably less than two blu-ray's in size and maybe slightly more in weight.  No justification for the astronomical shipping costs on it.

@amspire - I looked at the HP 3455A, but my primary resistance to buying it, though it does have some interest to me is that I'm primarily interested in collecting gear that I don't already have, and assuming I can fix my Agilent 34401a which is a 6.5 bench DMM, what are the advantages to buying the 3455A?  Granted the price at this moment is very good, (assuming I could buy it at the bid price and not the buy it now price) but I'm not sure whether it would add much value to the equipment I already own other than just being something else to take up space.  Please enlighten me and feel free to throw any more interesting projects my way.

@amspire - What's the value of having a pulse meter if I already own a sweep/function generator?  The unit I own goes from 0 - 2 Mhz not much but I assume suffient for the time being.

@kripton2035 - The ESR meter schematic I referred to earlier came from your site (well I got the link from your site based on another posting in these forums).  I'm still hoping you can figure out a 6 component ESR meter.  Though I assume with that few components it will be an analog meter and not a digital meter which i what I'm more interested in.  Keep me posted.

[alm]

@grenert - I looked at the ESR meter you recommend, but the 44 euros they charge to ship that to the US kills any sort of deal.  I imagine that thing probably less than two blu-ray's in size and maybe slightly more in weight.  No justification for the astronomical shipping costs on it.

Did you actually check the prices for shipping from Portugal to the US? I haven't checked either, but I know that shipping costs from Europe to the US tend to be much more expensive than the other way around, especially if it doesn't fit the dimensions to be considered a letter.

@amspire - I looked at the HP 3455A, but my primary resistance to buying it, though it does have some interest to me is that I'm primarily interested in collecting gear that I don't already have, and assuming I can fix my Agilent 34401a which is a 6.5 bench DMM, what are the advantages to buying the 3455A?  Granted the price at this moment is very good, (assuming I could buy it at the bid price and not the buy it now price) but I'm not sure whether it would add much value to the equipment I already own other than just being something else to take up space.  Please enlighten me and feel free to throw any more interesting projects my way.

You always need multiple meters . Because of its age, its functions to weight/volume ratio is fairly low. It won't do current measurement without external shunt, for example, and it's quite large and heavy. As most older equipment. You probably would have a better chance of repairing, though. You have a parts donor available. They tend to use less custom silicon. Documentation from that age tends to be much better, including HP Journal articles describing the design process. It would probably more of a repair project than an essential tool, although it's still a fine high-precision DMM.

@amspire - What's the value of having a pulse meter if I already own a sweep/function generator?  The unit I own goes from 0 - 2 Mhz not much but I assume suffient for the time being.

Higher repetition rates, variable rise/fall times, variable duty cycle and external triggering/gating. By no means an essential tool like a DMM, scope and function generator, but a nice tool none the less.