|
Post by boomzilla on Jun 8, 2020 8:06:43 GMT -5
I measured the speaker wires with the VOM: R+ = 0.10 Ohms R- = 0.15 Ohms L+ = 0.01 Ohms L- = 0.01 Ohms So amps would see 8.25 ohms on the (weaker) right speaker to 8.02 ohms on the (louder) left one. Is that extra 0.23 Ohms enough to make an audible difference? It's only a 2.9% difference between the two speaker cables... What are you using to get that kind of accuracy with low resistance measurements? My wife has decided that ‘my’ Digital VOM belongs in the downstairs tool closet, not at my upstairs workbench. I’m thinking the solution is to get something better, as that one’s not so great. I"m currently using a cheap Chinese (Vietnamese?) unit from Harbor Freight. But it's been reliable, so I have little incentive to upgrade it. If I wanted to be upscale, I'd buy a Fluke.
|
|
|
Post by boomzilla on Jun 8, 2020 8:08:48 GMT -5
VERY rare, but I’ve seen out of phase drivers (that were not intended to be) within the same cabinet. I'm convinced that this isn't that. Seems to be more of a break-in issue? The right speaker was just slightly stiffer in the suspension than the other, and also had the speaker cable with 10% more resistance. The combo made the right speaker audibly quieter. Once I changed wires & forced the cone to move a bit, the differences disappeared.
|
|
|
Post by AudioHTIT on Jun 8, 2020 10:27:39 GMT -5
What are you using to get that kind of accuracy with low resistance measurements? My wife has decided that ‘my’ Digital VOM belongs in the downstairs tool closet, not at my upstairs workbench. I’m thinking the solution is to get something better, as that one’s not so great. I"m currently using a cheap Chinese (Vietnamese?) unit from Harbor Freight. But it's been reliable, so I have little incentive to upgrade it. If I wanted to be upscale, I'd buy a Fluke. Mine is also said cheap Harbor Freight unit, but the lowest reading I can get (holding the probes together) is 0.3 ohm, maybe 0.2 if I rub them together to break through some tarnish. I’ve read of techniques to improve resolution, are you doing anything special or just reading the meter?
|
|
|
Post by boomzilla on Jun 8, 2020 10:56:49 GMT -5
OK - Maybe mine ISN'T a Harbor Freight... I'm just reading the meter for two digit precision. My unit looks a lot like this one, but without the brand marking:
|
|
KeithL
Administrator
Posts: 9,928
|
Post by KeithL on Jun 8, 2020 12:08:34 GMT -5
Normal low-end digital meters are NOT intended for taking accurate resistance measurements below about 1 Ohm. (You'll see that if you read the specs.)
- the resistance of the probe cables isn't that low - the resistance of the internal contacts, and especially the selector switch, isn't that low - and, as you've noticed, the resistance between the probes and what you're measuring isn't that low, and tends to change a lot - you'll also see that the accuracy is always rated as xx% +/- xx digits...
(so, for example, if you're reading ".003", and your meter accuracy is rated at xx% +/- 3 digits, your reading is totally meaningless)
In order to take accurate measurements of low-value resistances you really want a meter that can take "four wire Ohms measurements". Most higher-end bench-top multimeters include that option.
Or, if it's something you need to do, you can build a low-cost adapter that will allow you to do it with a cheap meter...
(It will let you read down to 1/1000 Ohm or so with your cheap meter... it's not difficult... it's just not something most people need to do.)
I"m currently using a cheap Chinese (Vietnamese?) unit from Harbor Freight. But it's been reliable, so I have little incentive to upgrade it. If I wanted to be upscale, I'd buy a Fluke. Mine is also said cheap Harbor Freight unit, but the lowest reading I can get (holding the probes together) is 0.3 ohm, maybe 0.2 if I rub them together to break through some tarnish. I’ve read of techniques to improve resolution, are you doing anything special or just reading the meter?
|
|
|
Post by ttocs on Jun 8, 2020 12:37:50 GMT -5
For resistances down to 0.01Ω I use this:
|
|
|
Post by AudioHTIT on Jun 8, 2020 14:26:41 GMT -5
For resistances down to 0.01Ω I use this: So that just does resistance? It seems well respected, I see it fully assembled and with specs here on Amazon. www.amazon.com/Blue-ESR-Tester-Fully-Assembled/dp/B00O0BMIYQNormal low-end digital meters are NOT intended for taking accurate resistance measurements below about 1 Ohm. (You'll see that if you read the specs.) - the resistance of the probe cables isn't that low - the resistance of the internal contacts, and especially the selector switch, isn't that low - and, as you've noticed, the resistance between the probes and what you're measuring isn't that low, and tends to change a lot - you'll also see that the accuracy is always rated as xx% +/- xx digits...
(so, for example, if you're reading ".003", and your meter accuracy is rated at xx% +/- 3 digits, your reading is totally meaningless)
In order to take accurate measurements of low-value resistances you really want a meter that can take "four wire Ohms measurements". Most higher-end bench-top multimeters include that option.
Or, if it's something you need to do, you can build a low-cost adapter that will allow you to do it with a cheap meter...
(It will let you read down to 1/1000 Ohm or so with your cheap meter... it's not difficult... it's just not something most people need to do.)
Since I want a second meter, and don't mind spending a little more, do you have a recommendation? My main goal is not precision circuit work or component selection, but making sure the cables I build have good solder joints or measuring a cable's total resistance. I occasionally check capacitance, just to see what it is. I also use it for various AC & DC Voltages, so a new digital multimeter appeals to me more than the specialized unit above. Here's a B&K that claims resistance measurements from 0.1 pF to 20 mF, 0.1 to 20 MΩ, I could live with that range (Amazon has it for a little less), do you know of another < $100 with similar or better resistance specs? Edit: I see that's not a multi-meter but a component tester. www.testequity.com/product/BAndK-Precision-815-Component-Tester-18637-1?gclid=Cj0KCQjww_f2BRC-ARIsAP3zarEVLcbySoa2mKpAoMSGOfYf_Oaza_G4dNxYcFWV4bxZqUqYQnFrgb0aAjepEALw_wcB
|
|
|
Post by ttocs on Jun 8, 2020 14:53:07 GMT -5
Since I want a second meter, and don't mind spending a little more, do you have a recommendation? My main goal is not precision circuit work or component selection, but making sure the cables I build have good solder joints or measuring a cable's total resistance. I occasionally check capacitance, just to see what it is. I also use it for various AC & DC Voltages, so a new digital multimeter appeals to me more than the specialized unit above. Here's a B&K that claims resistance measurements from 0.1 pF to 20 mF, 0.1 to 20 MΩ, I could live with that range (Amazon has it for a little less), do you know of another < $100 with similar or better resistance specs? Edit: I see that's not a multi-meter but a component tester. www.testequity.com/product/BAndK-Precision-815-Component-Tester-18637-1?gclid=Cj0KCQjww_f2BRC-ARIsAP3zarEVLcbySoa2mKpAoMSGOfYf_Oaza_G4dNxYcFWV4bxZqUqYQnFrgb0aAjepEALw_wcBIn the past I've tinkered with old guitar amps with the antique-iest being a 1951 Gibson. Being a fan of Uncle Doug I bought this ESR meter he recommended for checking in-circuit capacitors, but it can also double as a very low resistance meter - below 99Ω. And, like most meters which are capable of this, shorting the probes will zero out the meter thus testing only the component. I bought the assembled version. My go-to meter is a Fluke 87 I bough 25 years ago, and still works great. Back then there was only cheap and expensive, or surplus. I made the right call on this one. I also bough a cheap LCR meter for checking Capacitance and Inductance of the cables I've made. Good enough to have fun with for only $32
|
|
|
Post by boomzilla on Jun 9, 2020 2:15:04 GMT -5
Normal low-end digital meters are NOT intended for taking accurate resistance measurements below about 1 Ohm. (You'll see that if you read the specs.)
- the resistance of the probe cables isn't that low - the resistance of the internal contacts, and especially the selector switch, isn't that low - and, as you've noticed, the resistance between the probes and what you're measuring isn't that low, and tends to change a lot - you'll also see that the accuracy is always rated as xx% +/- xx digits...
(so, for example, if you're reading ".003", and your meter accuracy is rated at xx% +/- 3 digits, your reading is totally meaningless)
In order to take accurate measurements of low-value resistances you really want a meter that can take "four wire Ohms measurements". Most higher-end bench-top multimeters include that option.
Or, if it's something you need to do, you can build a low-cost adapter that will allow you to do it with a cheap meter...
(It will let you read down to 1/1000 Ohm or so with your cheap meter... it's not difficult... it's just not something most people need to do.)
In theory, KeithL, you're totally right. But in practice, I've found that even consumer VOMs (although they may not be ultimately accurate at less than one ohm) are NOT "totally meaningless." Within the meter's own inaccuracy, the "more or less" resistance measurement is not random, despite the meter's tolerance. In other words, yes, a meter may read anywhere between +/- some resistance - BUT it is consistently the SAME offset, regardless of its inaccuracy. For example, a meter that reads +0.2 Ohms at resistances below an ohm will CONSISTENTLY read +0.2 Ohms at ALL readings below an ohm. This means that If I want to compare two items, both of which are less than one ohm, I CAN use the meter and it will reliably tell me the difference between the two items - and with accuracy! I just can't rely on the measurements to be exactly accurate (despite their relative accuracy). To illustrate this, let's say I have a VOM that consistently reads +0.1 Ohm at one ohm or below measurements. I measure the first item (actual resistance = 1 Ohm) and get a reading of 1.1 Ohms. I then measure the second item (actual resistance = 0.5 Ohms) and get a reading of 0.6 Ohms. Despite being inaccurate ultimately (all readings are +0.1 ohms), the difference between the two measurements (0.5 Ohms) is spot-on! This is the same as using an inaccurate SPL meter (say, an iPhone) to read SPLs. It may not be accurate ultimately, but it will be consistent within its own measurements. Your dismissal of anything less than a four-wire meter for reading resistances of less than one ohm is a gross disservice to generations of electronics hobbyists, who have used consumer-grade VOMs to complete millions of successful projects. Were I a manufacturer evaluating components that would go into thousands of products, or a laboratory technician calibrating sensitive equipment, then the accuracy of a four-wire meter would be justified. But for hobbyists, your recommendation on this topic is totally right - but not at all helpful. Cordially - Glenn
|
|
|
Post by trevordj on Jun 9, 2020 2:25:01 GMT -5
Do you have access to a measurement microphone? A quick acoustic measurement would likely answer a lot of questions.
|
|
|
Post by Ex_Vintage on Jun 9, 2020 22:18:42 GMT -5
Precision - Shooting 5 bullets at a target and all bullets are 3 inches left and 2 inches low. Accuracy - all bullets are at the intended bullseye. You can bet that an instrument that is not accurate also has a general issue with repeatability. I agree with what Boom says with respect to a hobbyist, but the results you are getting are more of a "feel good" result than the actual circuit condition. If you want accuracy below say 5 ohms, a 4 wire measurement is required (and not just a 2 wire system with an adjustable offset), if you what a non quantitative, relative measurement, a consumer VOM is acceptable.
|
|
|
Post by boomzilla on Jun 10, 2020 21:48:06 GMT -5
Precision - Shooting 5 bullets at a target and all bullets are 3 inches left and 2 inches low. Accuracy - all bullets are at the intended bullseye. You can bet that an instrument that is not accurate also has a general issue with repeatability. I agree with what Boom says with respect to a hobbyist, but the results you are getting are more of a "feel good" result than the actual circuit condition. If you want accuracy below say 5 ohms, a 4 wire measurement is required (and not just a 2 wire system with an adjustable offset), if you what a non quantitative, relative measurement, a consumer VOM is acceptable. You GROSSLY underestimate the accuracy of consumer test equipment. Have you looked at the specs of current VOMs? Laser trimming of surface-mount resistors (as opposed to discreet components in older VOMs) has added orders of magnitude accuracy. From this website: themotorsmasters.com/best-multimeter/comes the article "10 best multimeters of 2020." The winner (Fluke 179) costs $323 on Amazon and has a resistance measurement accuracy of +/- 0.9% Yet the Extech MN35 ($22 on Amazon or about 1/15 the cost of the Fluke) has a resistance measurement resolution of 0.1 mOhms at +/- 0.8% variability So either the manufacturers are lying (which I doubt they'd do about test equipment) or you (and KeithL) are both wrong. Accuracy within a percent or two is obviously available from many consumer meters. And the absolute accuracy (measured as Ohms rather than %) gets BETTER as resistance falls, not worse. Are these meters as good as a 4-wire test rig? They are absolutely NOT. But for home use, they are incredible overkill in terms of accuracy. I further strongly disagree with your claim that a meter consistently off by X also has repeatability issues. My experience strongly contradicts your assertion. If I get a 0.03 Ohm reading and measure five times, my meter gives me from 0.02 to 0.04 EVERY time, and taking the average of multiple measurements ALWAYS reverts to the mean. You claim repeatability issues? Prove it. For laboratory calibration or production design, yes, better meters are available and are financially justified. But for home use, even cheap multimeters can be accurate to two decimal places, which is sufficient accuracy with room to spare.
|
|
|
Post by boomzilla on Jun 10, 2020 22:07:55 GMT -5
Do you have access to a measurement microphone? A quick acoustic measurement would likely answer a lot of questions. I do. The right speaker (EVERY speaker pair that I've measured in this room) is consistently about 1 to 1.25 dB lower in amplitude from the listening position than the left one. The room acoustics, combined with the higher resistance Right speaker wire combined to make the difference highly audible. My "pro" amp has individual linear input pots with notches so I can balance the channels (Right channel = +1 notch above the left channel). I normally run the input pots on the pro amp significantly less than 50% anyway so that the preamp can operate in its most linear part of the volume pot. So things work out in the end.
|
|