I know nothing about electronics, and in another forum:
http://www.diyaudio.com/forums/subwoofers/186704-driver-ts-crosscheck.html#post2542298
there was suggested to use a low-ohm meter to measure speaker driver Re as you would then get more reliable results.
There was one suggested (Low Ohm Meter - Measures 0.001 up to 1.999 Ohm), but its range is only 0.001 - 1.999 ohms, is that range common for low-ohm meters? I guess drivers have higher ohm values as well? You would also need a reliable ammeter to calibrate the device. There was also some issues for causion (on a native forum) about thermal effects, etc.
There was also suggested these parts:
Kjell & Company - Nordens bredaste sortiment av tillbehr fr hemelektronik
Kjell & Company - Nordens bredaste sortiment av tillbehr fr hemelektronik
(in swedish, but I guess you will know what it is anyway)
which I don't really understand what you could do with for my inteded purpose.
The good thing about the low-ohm meter is its four point measuring for example. A bench multi meter has also been suggested, but then I guess that prices really go up for reliable ones?
It got much more complicated than I first expected. I intend further on start building power amp. kits, but would I be better served just go measuring Re with a DMM?
http://www.diyaudio.com/forums/subwoofers/186704-driver-ts-crosscheck.html#post2542298
there was suggested to use a low-ohm meter to measure speaker driver Re as you would then get more reliable results.
There was one suggested (Low Ohm Meter - Measures 0.001 up to 1.999 Ohm), but its range is only 0.001 - 1.999 ohms, is that range common for low-ohm meters? I guess drivers have higher ohm values as well? You would also need a reliable ammeter to calibrate the device. There was also some issues for causion (on a native forum) about thermal effects, etc.
There was also suggested these parts:
Kjell & Company - Nordens bredaste sortiment av tillbehr fr hemelektronik
Kjell & Company - Nordens bredaste sortiment av tillbehr fr hemelektronik
(in swedish, but I guess you will know what it is anyway)
which I don't really understand what you could do with for my inteded purpose.
The good thing about the low-ohm meter is its four point measuring for example. A bench multi meter has also been suggested, but then I guess that prices really go up for reliable ones?
It got much more complicated than I first expected. I intend further on start building power amp. kits, but would I be better served just go measuring Re with a DMM?
Last edited:
Yes that is a normal range for low-ohm meters. They are not appropriate for measuring speakers, they are intended for things like measuring resistance of earth cabling etc.
All you need to do with your meter is short the probes to obtain the lead resistance then subtract this from your reading.
Note that in the thread you linked to, Ron suggested BUILDING a low-ohm measuring device using a current source, not an actual low-ohm meter.
All you need to do with your meter is short the probes to obtain the lead resistance then subtract this from your reading.
Note that in the thread you linked to, Ron suggested BUILDING a low-ohm measuring device using a current source, not an actual low-ohm meter.
Last edited:
Using a current source and calibrating your DMM & source with a high accuracy resistor, gives very good amateur results at very low cost.
A "normal DMM set to lowest resistance measures 199.9ohm at full scale.
The resolution of this scale is 0r1. Not nearly good enough for low ohm measurements.
With a range of 10mA and 100mA and 1A CCSs, you can use your 199.9mVdc scale to measure to milli-ohms (0r00x) accurately.
A "normal DMM set to lowest resistance measures 199.9ohm at full scale.
The resolution of this scale is 0r1. Not nearly good enough for low ohm measurements.
With a range of 10mA and 100mA and 1A CCSs, you can use your 199.9mVdc scale to measure to milli-ohms (0r00x) accurately.
Last edited:
Heres my PRACTICAL METHOD 1. you need to get a constant current diode, ccd 2. let say you get a nominal 1.8ma ccd 3. connect it up to a 9v supply and measure the current flow, say its 1767uA 4. now connect in series with ccd, your driver and measure the voltage across it 5. Re=Vdriver/1767uA (check the accuracy of your results by measuring a 10R 0.1% resistor)
Last edited:
I'm not sure I understand the reasoning behind the high tech resistance measurements.
On a 5 ohm copper coil at 20C, you get 5.4 ohms at 40C. If you can resolve to .1 ohm with your measurement you're doing pretty good.
Getting the accuracy from a regular DMM on the lowest range will probably require that you zero the leads before you measure.
.1 Ohm will not significantly affect the response of an 8 ohm nominal speaker unless it's in a bandpass box or something, and even then, it's going to go out of tune by more than that if you run it to max power.
On a 5 ohm copper coil at 20C, you get 5.4 ohms at 40C. If you can resolve to .1 ohm with your measurement you're doing pretty good.
Getting the accuracy from a regular DMM on the lowest range will probably require that you zero the leads before you measure.
.1 Ohm will not significantly affect the response of an 8 ohm nominal speaker unless it's in a bandpass box or something, and even then, it's going to go out of tune by more than that if you run it to max power.
Constant current methode works a treat
LM317 and one resistor to set the constant current
Drawing in the LM317 spec sheet
with 100 mA I have measured a batch of MPC74 0.22 homs and go the meter reading volt across them in milli volts.
There is here a tread about a LCR meter designed for speakers that may be worth looking at
An awesome ZLCR meter
How do you post links to treads is what get me cornered
so if you could help me on how to do this I will find it for you
LM317 and one resistor to set the constant current
Drawing in the LM317 spec sheet
with 100 mA I have measured a batch of MPC74 0.22 homs and go the meter reading volt across them in milli volts.
There is here a tread about a LCR meter designed for speakers that may be worth looking at
An awesome ZLCR meter
How do you post links to treads is what get me cornered
so if you could help me on how to do this I will find it for you
This thread came from a discussion about crosschecking T/S measurements. I suggested a (low tech) way to build an add-on to get more precision in DCR measurements (helps with inductors too
) Some of my inexpensive meters bounce ~+/- 0.2 ohms or so with the leads shorted, so what should I subtract?I wouldn't suggest buying a milliohmmeter unless you have another use for it.
Don't use the ohm-meter function unless there is absolutely no other way to get your measurement. It's really a very last resort.
Use the voltmeter function. One of the scales of the voltmeter will be the most accurate of all the functions with the DMM. It might be 1.999Vdc or just possibly 199.9mVdc, check the specification.
Use a current and voltage to measure resistances accurately.
with my cheapo DVMs substracting the shorted leads reading works very well.
Even bouncing 0.2 ohms would result in less than 5% error of a say 5.6 ohm Rdc.
There is no reason for better accuracy. 'Accurate' TS-measurements are an illusion. You would only lie to yourself.
For example try to measure TS specs at different room temperatures and be surprised how rubber surrounds react on a few degrees of enviromental temperatur changes.
Thiele-Small parameters are small signal parameters only valid at a given operating point. As soon the driver is playing some serious level, the voice coil heats up, suspension changes etc.
@buggson:
I suggest testing your DVM on a 2.2 and 5.6 ohm resistor or some similar values. 5% 2-5W metalfilm or MOX resistors are cheap and easy to get.
regards
Even bouncing 0.2 ohms would result in less than 5% error of a say 5.6 ohm Rdc.
There is no reason for better accuracy. 'Accurate' TS-measurements are an illusion. You would only lie to yourself.
For example try to measure TS specs at different room temperatures and be surprised how rubber surrounds react on a few degrees of enviromental temperatur changes.
Thiele-Small parameters are small signal parameters only valid at a given operating point. As soon the driver is playing some serious level, the voice coil heats up, suspension changes etc.
@buggson:
I suggest testing your DVM on a 2.2 and 5.6 ohm resistor or some similar values. 5% 2-5W metalfilm or MOX resistors are cheap and easy to get.
regards
My sears 82140 DVM ($29) will resolve 0.1 ohm but won't actually go to zero. What the low number is has a lot to do with how recently the battery was changed. Besides shorting the lead, I measure a 5% 1 ohm resistor before testing under 10 ohm. Then I add or subtract the reading the meter has on the 1 ohm. You can also use 2 or 5 ohm resistors for reference if that is closer to what you are measuring.
I made exactly the same: but using a 78L24 voltage regulator, and set that to give a 10,0 ma current: I have two 79,6 ohm 0,5% resistors, and could find the appropriate resistors: 7.500//5.100 ohms gives a better than 1% precision. It measures as low as the DVM goes, in practice 0,1 ohm is no problem.
albert
Hi Doc,
Actually I myself just bought a new meter, 4 1/2 digits, VC8045 bench-top, mains-fed DVM, that has a resolution down to 0,01 ohm on the 200 ohm range. I was not satisfied with my own breadboard.
It cost me $ 99 plus shipping and tax; imho a bargain. Others have a lowest range of 600 with the same resolution, but have a 50-100% higher price tag.
I have seen meters with four pole plugs (that is two current emitting and two voltage sensing) meters, that overcome the problem of the leads themselves being measured.
Why there is not a kind of 'tara' function, like on the common household weight, I do not understand. Short the leads, press the tara button, and the setting becomes zero, the resistance is then shown as-is.
Look, I am inventing here on the spot!!!
albert
Actually I myself just bought a new meter, 4 1/2 digits, VC8045 bench-top, mains-fed DVM, that has a resolution down to 0,01 ohm on the 200 ohm range. I was not satisfied with my own breadboard.
It cost me $ 99 plus shipping and tax; imho a bargain. Others have a lowest range of 600 with the same resolution, but have a 50-100% higher price tag.
I have seen meters with four pole plugs (that is two current emitting and two voltage sensing) meters, that overcome the problem of the leads themselves being measured.
Why there is not a kind of 'tara' function, like on the common household weight, I do not understand. Short the leads, press the tara button, and the setting becomes zero, the resistance is then shown as-is.
Look, I am inventing here on the spot!!!
albert
Hi back, Albert. Took a look at the VC8045 and in my opinion it is indeed probably a bargan at $99. Over the years I've used quite a few four terminal equipped bench meters and even at one time had a Datron 1082 8.5 digit DMM resident on my bench (along with a maxed out Tek2465 scope). As to the tara function, it's already incorporated in many DVM's (including going back to that Datron 1082). Both my Fluke 87 and my HP handheld have it. I know HP (now agilent) incorporated it on some of their meters as far back as the seventies.
Anyway, I can atest from my cal lab days that virtually all high precision ohms measurement are made using a 4 wire system.
Doc
Anyway, I can atest from my cal lab days that virtually all high precision ohms measurement are made using a 4 wire system.
Doc
Maxim App Note 106 works well
If you're looking for a good general purpose milliohm meter I've used the circuit from Maxim Application Note 106 with good results. Quite an elegant circuit that will work with any sensible reference, op-amp and low-Rds MOSFET. I'm still on my original batteries after 2-3 years use.
If you substitute a dual op-amp you can wire the other half as a comparator to light an LED when the FET gate voltage gets near the +9V rail - ie. Over-range indicator (resistance too high to maintain constant current).
I can't find the App note on the Maxim site anymore but there are plenty of hits on a Googe search, eg. http://www.wentztech.com/radio/Projects/Projects_assets/AN106.pdf
If you're looking for a good general purpose milliohm meter I've used the circuit from Maxim Application Note 106 with good results. Quite an elegant circuit that will work with any sensible reference, op-amp and low-Rds MOSFET. I'm still on my original batteries after 2-3 years use.
If you substitute a dual op-amp you can wire the other half as a comparator to light an LED when the FET gate voltage gets near the +9V rail - ie. Over-range indicator (resistance too high to maintain constant current).
I can't find the App note on the Maxim site anymore but there are plenty of hits on a Googe search, eg. http://www.wentztech.com/radio/Projects/Projects_assets/AN106.pdf
5% is probably about as accurate as t/s measurements can get, but that is no reason to settle for low accuracy measurements. for a nominal 4 ohm driver, the error increases to 10%, and for a 2 ohm, 20%. There are many other sources of error in t/s measurement and this is a simple one to eliminate. Next is to decide whether Re=Rdc or not, for inductive drivers, probably not.
Also for inductive drivers the shape of the peak changes...and the FR around resonance...- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.