The easiest way is to compare it against a meter you know is a good. I have an old Fluke 75 meter and have burned out the fuse resistor twice so far. Its only a few millivolts off from the calibrated HP 34401 meters at work. Do you know anyone with access to calibrated equipment?
DMMs will start to disagree with a scope when the sinewave has distortion or stops being a sine. Frequency response plays a big part in this too.
DMMs will start to disagree with a scope when the sinewave has distortion or stops being a sine. Frequency response plays a big part in this too.
Thanks for reply. I don't know anyone with calibarted equip. I live in a small town. Mabye it is in spec It could be the wave forms like you said. I do have a agilent power supply which is supposed to be a very high end supply. Could this be accurate enough to do a cal or would it vary like most everything else?
If you don't have access to calibrated equipment your best bet will be to calibrate to a band gap reference or failing that the Vf on an LED. If you averaged a green, red & yellow LED I dare say you'd be within ±5%.
Or are you looking to calibrate the 500V scale?
AC is a different matter of course, as astouffer mentions.
I think you can do everything with a good volt reference. Anyone know other good references? What where those battery cells they used to use before NIST?
Or are you looking to calibrate the 500V scale?
AC is a different matter of course, as astouffer mentions.
I think you can do everything with a good volt reference. Anyone know other good references? What where those battery cells they used to use before NIST?
exactly.
The forward volt drop on an LED is very stable and consistent. That's why it make such a good reference for current sources. Don't want to quote voltages from memory because i'd be wrong but they're all in the 1-2V range (except blue of course)
Knowing the rms behaviour of your meter you could use this to calibrate AC volts also.
The forward volt drop on an LED is very stable and consistent. That's why it make such a good reference for current sources. Don't want to quote voltages from memory because i'd be wrong but they're all in the 1-2V range (except blue of course)
Knowing the rms behaviour of your meter you could use this to calibrate AC volts also.
You can buy a reference grade voltage reference like an LM4030, LM4132, MAX6033, REF5010, etc. But even those are typically 0.01% at best and your Fluke is supposed to be more accurate than that. I have a high-end Agilent bench DMM with a 4 figure price tag that's still under factory calibration and reads to 7 digits. But my ancient Fluke 87 agrees closely enough with it for 99% of anything I need to measure. And the Fluke has been dropped more times than I can count. So, unless you have reason to believe your Fluke is wrong, I'd just trust it.
To Rocket scientist I was thinking there might be a problem due to some measurements i was taking the other day in the 350-400mv ranfge. The volt kept disapearing around 360mv & up. It could have been just an unstable source but I was just thinking meter may have been out of cal & I was actually in a higher volt range with source causing it to be unstable..
gto127 said:
Yeah, regulators are not even close to a DMM in accuracy. Nor are zeners or most anything else you can think of including LEDs. Most vary by several percent or more from the factory and also with changes in temperature. Voltage references were made for the purpose.
Which brings up another point. If you do decide to use a voltage reference you'll want to calibrate the DMM with the reference as close to its reference temperature as possible. References vary in their performance across temps, but it's safest and more accurate to use it at whatever temp is on the datasheet for the reference specs--most likely 20C or 25C.
Hi gto127,
The 87's do in fact hold their calibration pretty well, however any change in case dimensions or PCB position will throw out the high frequency AC voltage scales. Your crest factor on that meter is quite good, I can't remember exactly but I think it's over 5 and true RMS. Certainly the CD servos will not tax the high frequency response at all. I would wonder if the RMS function tracks the peak values though.
The best "at home" suggestion are some of the newer, high accuracy voltage references. They should be left running for a couple hours to settle in first. These will be more stable than LEDs or anything else. They should be low noise as well. I would highly recommend having your meter professionally calibrated at this point though. The other voltages and AC ranges will be next to impossible to reliably confirm. The current DC cal may easily be closer than anything you can do at home.
The cal lab should really be using a Fluke 5500A (multifunction calibrator) as a minimum. This allows a full calibration in 1/2 an hour ~ 45 min at the most. If they have a top cover drilled for the adjustment locations, your cal will be more accurate. I do happen to know that Transcat in Rochester has this (because I sent them a top cover when I worked there). Another good calibration lab is Teletek. I have dealt with him as well. Of course, you can always use Fluke. Can't beat them and they should have any parts required on hand.
After a proper calibration, you should be good for another 10 years! Just don't drop your meter. It probably won't cause it not to work, but dropping it will probably affect the calibration.
If you can find an HP 3478A meter, get one. It will probably need the battery changed (it holds the calibration constants), but if you can replace the battery before it loses calibration, you should be fine. You will need two batteries to pull and replace the internal battery to keep the BATT line powered. They are non-rechargeable Lithium batteries.
An HP 34401A is very hard to beat as the perfect bench meter, mine is my best friend. Those are expensive, even used though.
A final question for you. What are you doing with a CD player alignment that requires that accuracy? Most offset adjustments should be done with a 'scope set on DC (due to noise), laser current is 3 digit.
Hi RocketScientist,
-Chris
The 87's do in fact hold their calibration pretty well, however any change in case dimensions or PCB position will throw out the high frequency AC voltage scales. Your crest factor on that meter is quite good, I can't remember exactly but I think it's over 5 and true RMS. Certainly the CD servos will not tax the high frequency response at all. I would wonder if the RMS function tracks the peak values though.
The best "at home" suggestion are some of the newer, high accuracy voltage references. They should be left running for a couple hours to settle in first. These will be more stable than LEDs or anything else. They should be low noise as well. I would highly recommend having your meter professionally calibrated at this point though. The other voltages and AC ranges will be next to impossible to reliably confirm. The current DC cal may easily be closer than anything you can do at home.
The cal lab should really be using a Fluke 5500A (multifunction calibrator) as a minimum. This allows a full calibration in 1/2 an hour ~ 45 min at the most. If they have a top cover drilled for the adjustment locations, your cal will be more accurate. I do happen to know that Transcat in Rochester has this (because I sent them a top cover when I worked there). Another good calibration lab is Teletek. I have dealt with him as well. Of course, you can always use Fluke. Can't beat them and they should have any parts required on hand.
After a proper calibration, you should be good for another 10 years! Just don't drop your meter. It probably won't cause it not to work, but dropping it will probably affect the calibration.
If you can find an HP 3478A meter, get one. It will probably need the battery changed (it holds the calibration constants), but if you can replace the battery before it loses calibration, you should be fine. You will need two batteries to pull and replace the internal battery to keep the BATT line powered. They are non-rechargeable Lithium batteries.
An HP 34401A is very hard to beat as the perfect bench meter, mine is my best friend. Those are expensive, even used though.
A final question for you. What are you doing with a CD player alignment that requires that accuracy? Most offset adjustments should be done with a 'scope set on DC (due to noise), laser current is 3 digit.
Hi RocketScientist,
3458A ??! Or is it an Agilent 34410 or 11? The 3457A is very close to a 34401A (but slower readings) and a 3456A seems to be more stable, but maybe not as accurate as the 34401A.
-Chris
Anatech:
Im calibrating a Nak 0ms7E which has 10+ steps to cal & like you say I should be using a scope for most of it but its hard to tell where the lines are sometimes on the focus & tracking servos.(its hard to tell looking at these signals when you are at 50mv or 100mv exactly I have a older scope without digital readout but has div points) I was hoping to get it more accorate with Fluke. I'm also calibrating a Krell with CDM1MKII which requires 2 DVM measurements on the laser power & focus offset. The focus ofset was where I was getting the unstable readings.
Im calibrating a Nak 0ms7E which has 10+ steps to cal & like you say I should be using a scope for most of it but its hard to tell where the lines are sometimes on the focus & tracking servos.(its hard to tell looking at these signals when you are at 50mv or 100mv exactly I have a older scope without digital readout but has div points) I was hoping to get it more accorate with Fluke. I'm also calibrating a Krell with CDM1MKII which requires 2 DVM measurements on the laser power & focus offset. The focus ofset was where I was getting the unstable readings.
Hi gto127,
Okay, for that you only require a 'scope. You need to place the fuzz on a dividing line. Most meters will not indicate correctly, even the "RMS" responding units tend to work on the positive envelope. Therefore you can not even use a meter to begin with.
You do not need cursors at all. 50 mV/Div should do it. Almost any 'scope is up to this job, your sweep will be a little faster than when it flickers, too fast a sweep will be useless. All you need to do is to confirm your zero at the graticule you want to use, then cal using a DC source for 50 and 100 mV. Any of those voltage reference ICs will do nicely once divided down (confirm with your Fluke). Use that as a known voltage source to set your 50 or 100 mV span as required.
On the Krell CD, you only require 3 decimals to be accurate, something your Fluke will be fine for. The laser current from the factory is only measured to 3 figures to begin with. Anything more demanded beyond that tells me that someone at Krell may have a Napoleon complex going on. This will vary a bit with temperature as well, making highly accurate measurements simply stupid. For the focus offset measurement, this will also depend on the disc table height. Some tables drop down over time (Philips), it varies with make. Your 'scope will provide a very close measurement here. Of course, you can use your Fluke 87 here as well. I don't know what the exact procedure is.
-Chris
Okay, for that you only require a 'scope. You need to place the fuzz on a dividing line. Most meters will not indicate correctly, even the "RMS" responding units tend to work on the positive envelope. Therefore you can not even use a meter to begin with.
You do not need cursors at all. 50 mV/Div should do it. Almost any 'scope is up to this job, your sweep will be a little faster than when it flickers, too fast a sweep will be useless. All you need to do is to confirm your zero at the graticule you want to use, then cal using a DC source for 50 and 100 mV. Any of those voltage reference ICs will do nicely once divided down (confirm with your Fluke). Use that as a known voltage source to set your 50 or 100 mV span as required.
On the Krell CD, you only require 3 decimals to be accurate, something your Fluke will be fine for. The laser current from the factory is only measured to 3 figures to begin with. Anything more demanded beyond that tells me that someone at Krell may have a Napoleon complex going on. This will vary a bit with temperature as well, making highly accurate measurements simply stupid. For the focus offset measurement, this will also depend on the disc table height. Some tables drop down over time (Philips), it varies with make. Your 'scope will provide a very close measurement here. Of course, you can use your Fluke 87 here as well. I don't know what the exact procedure is.
-Chris
anatech said:
I have the newer 34410A which I mainly purchased for audio use as it does quite a few things better in that area than a handheld portable DMM can. I also like to use the 34410A for data acquisition where it can be monitored remotely via LAN or internet. Fluke does a great job with handheld DMMs (although I haven't tried the new Agilent models) but, IMHO, Agilent offers better bench meters. Fluke is behind in that area.
For the focus offset measurement, this will also depend on the disc table height. Some tables drop down over time (Philips),
Yes this is a philips.CDM1mkII so it may not be the correct height. I have a borousch alignment disc. Could I use that & adjust to a certain focus ofset equivalent since the heigtht of the disc table is unknown?Thanks for the scope tips.
Yes this is a philips.CDM1mkII so it may not be the correct height. I have a borousch alignment disc. Could I use that & adjust to a certain focus ofset equivalent since the heigtht of the disc table is unknown?Thanks for the scope tips.
Hi RocketScientist,
I'm looking at an original 34401A. It was originally purchased for two reasons, the main reason was audio servicing. The second was design work that required either both the precision and data acquisition features. I'm on the GPIB system, but I'm lusting after the LXI featured gear. A new 34411A (or 34410A) would make me an extremely happy camper.
Over the time I used it for service, the time it saved me has paid this meter back many times over. I know two other audio technicians who have one. The guy who bought his new fully appreciates the meter's abilities. The other guy does more work, but got his used in a deal. He likes it, but I don't think he really understands it very well.
Since mine is an ancient model compared to yours, I can tell you that it holds it's calibration extremely well. I had it certified once and optimized once (I did it at the lab). It didn't require adjustment really, but the DC offset was a touch off. I just got an older 3457A meter, certified from Teletek (Joe). It agrees with the 34401A. Amazing! The 3457A takes longer for a measurement, but that is because of the averaging. I leave it in this mode.
BTW, I highly recommend Teletek as a source for used and certified HP equipment. Other people may be less expensive, but the items Joe sells are fully functional and normally certified. Teletek is a calibration lab.
I bought three 3478A meters from three different vendors off Eeeekbay. All these meters needed work, sadly I'm not too bad with these now. I had to correct the cal one each, two had lost the calibration constants. I use the 34401A as a standard. One point should be made about these meters. The HP/Agilent 34401A and newer, plus the older HP 3456A, will take readings up to 1 KV. The HP 3478A and HP 3457A will only accept inputs up to 300 VDC. Tube guys need to take note of this. I suspect that most higher voltage readings our tube / valve members do not require the accuracy of the 34401A. If it is required, I guess a single range DVM using something along the lines of a Maxim ICL-7135 or newer MAX132 with a dedicated input divider and a quiet external reference would more than suffice (4 1/2 digits). I do plan to build a few of those in the future. I'd hate to blow up an HP just because I'm use to the 34401A ranges.
-Chris
I'm looking at an original 34401A. It was originally purchased for two reasons, the main reason was audio servicing. The second was design work that required either both the precision and data acquisition features. I'm on the GPIB system, but I'm lusting after the LXI featured gear. A new 34411A (or 34410A) would make me an extremely happy camper.
Over the time I used it for service, the time it saved me has paid this meter back many times over. I know two other audio technicians who have one. The guy who bought his new fully appreciates the meter's abilities. The other guy does more work, but got his used in a deal. He likes it, but I don't think he really understands it very well.
Since mine is an ancient model compared to yours, I can tell you that it holds it's calibration extremely well. I had it certified once and optimized once (I did it at the lab). It didn't require adjustment really, but the DC offset was a touch off. I just got an older 3457A meter, certified from Teletek (Joe). It agrees with the 34401A. Amazing! The 3457A takes longer for a measurement, but that is because of the averaging. I leave it in this mode.
BTW, I highly recommend Teletek as a source for used and certified HP equipment. Other people may be less expensive, but the items Joe sells are fully functional and normally certified. Teletek is a calibration lab.
I bought three 3478A meters from three different vendors off Eeeekbay. All these meters needed work, sadly I'm not too bad with these now. I had to correct the cal one each, two had lost the calibration constants. I use the 34401A as a standard. One point should be made about these meters. The HP/Agilent 34401A and newer, plus the older HP 3456A, will take readings up to 1 KV. The HP 3478A and HP 3457A will only accept inputs up to 300 VDC. Tube guys need to take note of this. I suspect that most higher voltage readings our tube / valve members do not require the accuracy of the 34401A. If it is required, I guess a single range DVM using something along the lines of a Maxim ICL-7135 or newer MAX132 with a dedicated input divider and a quiet external reference would more than suffice (4 1/2 digits). I do plan to build a few of those in the future. I'd hate to blow up an HP just because I'm use to the 34401A ranges.
-Chris
Hi gto127,
Essentially, any corrections you make should be only slight. So, do your focus offset, then observe the eye pattern and adjust that offset a little each way. If you need to make an adjustment to improve the eye pattern, the disc may not read when it is read the first time for the TOC data. In other words, your best eye pattern must occur when you make the electrical adjustments. If you are worried that someone may have adjusted the table height, take measurements from similar transports. They will be identical.
-Chris
Ahhhhh, watch out! There is a procedure that involves adjusting the thrust bearing in the bottom of the disc motor to correct the height. You really do not want to do this. There is but one reason for the table height to appear to be wrong. Sometimes the suspension for the laser head lens will sag. The fix is to replace the head, not to "correct" the disc table height. When I was doing warranty for Revox, I got bit by this issue (and a few others involving the tray assy in the newer one. I will always remember this lesson. It would have been nice if they had warned me though.
Essentially, any corrections you make should be only slight. So, do your focus offset, then observe the eye pattern and adjust that offset a little each way. If you need to make an adjustment to improve the eye pattern, the disc may not read when it is read the first time for the TOC data. In other words, your best eye pattern must occur when you make the electrical adjustments. If you are worried that someone may have adjusted the table height, take measurements from similar transports. They will be identical.
I'm not familiar with that test disc. The clear glass disc is to align the arm travel (don't even try!, it's a severe pain in the rear end!). The table height will be so many mm above the chassis surface. Record the one you have now. Ask to see if anyone else has one open they can measure for you.
-Chris
anatech said:
Yeah, for the rare HV stuff I do I use my Fluke 189 which is Cat III to 1000 volts (and probably well beyond that knowing Fluke). I think the 87 is as well although I don't have it in front of me.
An instrument I use more than I thought I would is my Fluke 192 Scopemeter. The inputs are also good to 1000 volts but the best part is they're ISOLATED--not just from the rest of the world, but also from each other. So it's very handy for doing stuff like putting it across emitter resistors in a running amplifier without having to worry about blowing anything up with grounding issues or buying expensive isolated differential probes that cost more than the Scopemeter. I would imagine one would be very handy for tube work (I personally don't do anything with tube gear).
LXI is great. I have an Agilent MSO 4+16 scope that's LXI as well and it's great to have it on the LAN. Agilent's software suite is kind of bloatware but it works fine. I also have a Tektronix AFG3022 dual channel arbitrary function generator that's LAN connected as well. LXI/LAN is much nicer than GPIB or USB as I can also use the PC's at my desk, not just the one at my test bench.
I couldn't justify getting the LXI modules for my dual Sorensen DCS80-15E 1.2KW power supplies (0-80V 0-15A each). They're mostly "set and forget" for any given project. I did, however, have to run a dedicated 20 amp circuit to my test bench to keep them happy working on an especially big amp project.
I used to have a Tek DPO3000 4 channel scope but I hated it. The screen resolution was bad, the screen updates were slow and the user interface drove me nuts. The Agilent is a dream by comparison. Tek is funny. Several years ago at a trade show they told me they would never make a mixed signal scope like the Agilent models because it was "too much of a compromise". Now they have several mixed signal scopes just trying to keep up with Agilent!
The Tek DPO/MSO4000 seems like it addresses a lot of the problems of the 3000 and has cool zoom and playback features. But the Agilent still has faster screen updates and catches more detail. It's kind of funny (and sad in a way) that Tek, once the undisputed leader in scopes, has been mostly playing catch up to Agilent for many years now. I have an old analog 100Mhz Tek I keep around for those rare times when a digital scopes won't do what I need. I also have a high resolution 16 bit Pico digital scope for audio work.
And, like you, I've had mixed luck with eBay and test equipment. My analog Tek was a used purchase and it has numerous problems. I've bought all my other expensive stuff new or a few pieces factory reconditioned but under full warranty. While Agilent gear isn't supposed to be heavily discounted when new, there are ways to get a fairly significant discount and still buy from an authorized dealer. Plus Agilent runs promotions now and then.
The only thing about Agilent I'm not crazy about is they're slapping their name on low-end stuff they didn't design--namely Rigol scopes, other Rigol products, and now some cheesy Chinese OEM power supplies. The stuff just isn't that great and nowhere near up to Agilent standards. Their handheld scopes are made by Rigol as are the DSO/MSO 1000 series. I kind of wonder if their new fancy handheld DMMs are their own design or another OEM/ODM product? I haven't wanted to find out. So I stick with Fluke in that area. They're hard to beat.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.