High end Meters and burden voltage

[Freighternut]

I am close to purchasing an Agilent 34410A bench meter.

In looking at the spec's for current I note a burden voltage of < 2.0 v on the 3 amp range!!!!

How can this be on such a high end meter, even the much cheaper 34405A has a burden voltage of < 0.6 Volt on the 10 amp range.

Can anybody enlighten me as to why the 34410A would have such a huge burden voltage?

[robrenz]

You may want to look at this thread. post #14 shows a meter with better burden votage
https://www.eevblog.com/forum/index.php?topic=5676.0

[amspire]

Can anybody enlighten me as to why the 34410A would have such a huge burden voltage?

At a guess, HP has taken the stance "We cannot properly protect against faults into a 0.1 ohm sense resistor, so you can't have one. If you want to measure large current get an external shunt."

I don't know what is in this meter but they may have put something like a PTC thermistor in series with the current range.

You can possibly protect a 0.1 ohm current sense resistor against accidental connections across the 120V or 240V mains, but I doubt you can protect against a large charged capacitor with 1000V DC on it. Even with the best fuse, the meter would be damaged as well, and HP are making the point that they are not going there with their precision instrument.

Richard

[alm]

It's a trade-off between noise floor at the lower ranges, number of shunt resistors and burden voltage at the higher ranges. Apparently they optimized for the former, and not for the latter. Burden voltage tends to be worse at the top of the range. Fluke incorporates a feedback ammeter in the Fluke 8508A, but this has not (yet) trickled down to the lower models.

I don't see the issue with protection, you just need to make sure the circuit can withstand the short circuit current at the max. rated voltage until the fuse blows. Bench meters often don't use HRC fuses and are not usually rated for high energy circuits, so I'm not sure if a large 1000V DC cap is even within specs.

[IanB]

You don't measure multi-amp currents by feeding them through the internals of a precision benchtop meter!

If you have a need for such measurements, get a precision external current shunt and use that. If you can afford a bench meter you can afford some proper shunt resistors. 

[Freighternut]

Thanks for the input. I guess the issues around protection is why Agilent push their external current shunt for higher currents.

 The 34330A shunt is good to 30 Amp and has a resistance of 0.001 ohm. Accuracy is quoted at
plus/minus 0.3% (DC - 1 kHz).

The Fluke meter looks great, but I have a number of reasons for staying with Agilent meters, not the least being that I have already shelled out for the Agilent 34405A. After looking closer at the spec's for the 34405A and having a demo model to play with while I wait for mine to arrive, I now believe that the 34410A will better suit my needs.

Thanks again for the input.

[amspire]

It's a trade-off between noise floor at the lower ranges, number of shunt resistors and burden voltage at the higher ranges. Apparently they optimized for the former, and not for the latter. Burden voltage tends to be worse at the top of the range. Fluke incorporates a feedback ammeter in the Fluke 8508A, but this has not (yet) trickled down to the lower models.

It is nothing to do with the noise floor. The 10mA range has a 0.03v burden voltage, and so if they could have the same for a 3A range.

They could have chosen to put an extra shunt resistor in.  I don't think they were saving money, they just don't want to go there at all. As I said, the message is - if you want to measure currents of an amp or more, use a current shunt. If a circuit uses 100mA or more, you just cannot use the Agilent meter to measure the circuit's supply currents using the meter's current ranges.

Very accurate 0.01 ohm resistors can be made, and the lower the resistance, the less power dissipated so Agilent are making things very hard for themselves by using a resistor of perhaps 0.5 ohms for measuring 3 Amperes. The power dissipated by a 0.5 ohm current sense resistor is 2500 times the power in a 0.01 ohm resistor, and to meet Agilent's accuracy specs, it will cost them a lot for that shunt resistor that can stay very accurate with that amount of thermal cycling.

If they wanted to, they could add any protection that is in their cheaper hand held meters, but they just don't want to touch that. The idea of a feedback ammeter is interesting since it means there is physical shunt resistor shorting the inputs. A great idea for precision measurement - not such a great idea for measuring practical currents from a circuits power supply that may have high frequency current pulses. I can understand Agilent may not want to go that way. Fluke may drop the idea as well if customer's hate it.

I don't see the issue with protection, you just need to make sure the circuit can withstand the short circuit current at the max. rated voltage until the fuse blows. Bench meters often don't use HRC fuses and are not usually rated for high energy circuits, so I'm not sure if a large 1000V DC cap is even within specs.

I don't have the cat I 1000V and cat III 300V specs, but I think that CAT1 is tested with a 30 ohm source, cat III with a 2 ohm source.

The Agilent meter does Cat III 300V so does that mean it has to be able to handle a 150Amp fault current into the current input? That is a really big ask for a precision instrument. People often think fuses are some kind of magical protection, but they are not. By the time a fuse has completed arcing on a 300V DC fault, lots of damage can be done to a circuit. AC is much easier as it cannot arc for more than half a mains cycle. I think you would find that the main reason for the high burden voltage is the protection they have added.

But I think that it is not a matter of specs. Agilent designers will be designing beyond that, and you know that if their new meter has a higher burden voltage then their older meter, it is an intentional choice they have made. If Fluke have a lower burden voltage, then Agilent would probably argue that Fluke have got it wrong.

Fluke is saying "we will give you a lower burden voltage but be careful not to wreck the meter" and Agilent is saying "use an external current shunt for an amp or more".

Richard

[amspire]

The 34330A shunt is good to 30 Amp and has a resistance of 0.001 ohm. Accuracy is quoted at
plus/minus 0.3% (DC - 1 kHz).

You can get current shunts with calibration accuracies up to 2ppm (0.0002% - costs about $4000 though), so with external current shunts, you can measure currents far more accurately then the current ranges of any multimeter.

You can also get 10ppm (0.001%) accurate 0.0001 ohm shunts, so you can measure up to hundreds of amperes with very high accuracy.

Basically you can get any accuracy you are prepared to pay for.

Richard

[alm]

It is nothing to do with the noise floor. The 10mA range has a 0.03v burden voltage, and so if they could have the same for a 3A range.

That would require either an extra shunt resistor for the 3A range, or it would further reduce the burden voltage in the 10mA range, thereby increasing the noise floor.

They could have chosen to put an extra shunt resistor in.  I don't think they were saving money, they just don't want to go there at all. As I said, the message is - if you want to measure currents of an amp or more, use a current shunt. If a circuit uses 100mA or more, you just cannot use the Agilent meter to measure the circuit's supply currents using the meter's current ranges.

I'm not sure. Why would they spec the meter to 3A but only make it work well up to 100mA? I agree that an external shunt is the correct solution for larger currents.

Very accurate 0.01 ohm resistors can be made, and the lower the resistance, the less power dissipated so Agilent are making things very hard for themselves by using a resistor of perhaps 0.5 ohms for measuring 3 Amperes. The power dissipated by a 0.5 ohm current sense resistor is 2500 times the power in a 0.01 ohm resistor, and to meet Agilent's accuracy specs, it will cost them a lot for that shunt resistor that can stay very accurate with that amount of thermal cycling.

24h accuracy spec is 0.1% at 3A, not that tight. About ten times worse than the low current ranges. Very low value resistors with high accuracy are also expensive. A 10 mOhm shunt is overkill since the wiring and contact resistance is likely going to swamp the shunt resistance.

The idea of a feedback ammeter is interesting since it means there is physical shunt resistor shorting the inputs. A great idea for precision measurement - not such a great idea for measuring practical currents from a circuits power supply that may have high frequency current pulses. I can understand Agilent may not want to go that way. Fluke may drop the idea as well if customer's hate it.

A physical shunt resistor shorting the inputs? What are you talking about? The difference between a shunt ammeter and a feedback ammeter is that the feedback ammeter replaces the shunt resistor by an op-amp with an almost zero burden voltage. You would not typically use it for 3A, however, only for the low ranges.

The Agilent meter does Cat III 300V so does that mean it has to be able to handle a 150Amp fault current into the current input?

I don't believe IEC61010 requires the equipment to remain in spec or survive, just handle the transient without blowing up.

[robrenz]

8846A for reference

Range      resolution    shunt          max burden voltage
100 uA     100 pA       100 Ohm     <0.015 V
1 mA        1 nA          100 Ohm     <0.15 V
10 mA      10 nA        1 Ohm         <0.025 V
100 mA    100 nA      1 Ohm         <0.25 V
400 mA    1 uA          1 Ohm         <0.50 V
1 A          1 uA          0.01 Ohm    <0.05 V
3 A          10 uA        0.01 Ohm    <0.15 V
10 A        10 uA        0.01 Ohm     <0.5 V

[amspire]

It is nothing to do with the noise floor. The 10mA range has a 0.03v burden voltage, and so if they could have the same for a 3A range.

That would require either an extra shunt resistor for the 3A range, or it would further reduce the burden voltage in the 10mA range, thereby increasing the noise floor.

To lower the 100mA and 3A burden voltages, it would needs an extra resistor, but Agilent are not trying to save the cost of an extra resistor. If it need extra inputs, like a 20A range on a multimeter, they can add those to a meter of this price. It does not make sense to suggest that they can add an extra low value shunt resistor in their $100 multimeters, but not in their expensive bench DMMs. A total additional cost of a few dollars only. They have just decided not to attempt to be a useful amp-level meter.

24h accuracy spec is 0.1% at 3A, not that tight. About ten times worse than the low current ranges. Very low value resistors with high accuracy are also expensive. A 10 mOhm shunt is overkill since the wiring and contact resistance is likely going to swamp the shunt resistance.

There is zero error from the wiring and contact resistance. The current shunt resistor has taps for the voltmeter, and that bypasses all  wiring and contact resistance.

The idea of a feedback ammeter is interesting since it means there is physical shunt resistor shorting the inputs. A great idea for precision measurement - not such a great idea for measuring practical currents from a circuits power supply that may have high frequency current pulses. I can understand Agilent may not want to go that way. Fluke may drop the idea as well if customer's hate it.

A physical shunt resistor shorting the inputs? What are you talking about?

I meant that the current shunt resistor is a low value resistor connected across the two multimeter inputs. Ideally, it should be low enough to look like a short to the circuit.

The difference between a shunt ammeter and a feedback ammeter is that the feedback ammeter replaces the shunt resistor by an op-amp with an almost zero burden voltage. You would not typically use it for 3A, however, only for the low ranges.

Yes - this idea has been around as long as opamps, so it is not a Fluke invention. It has always been used in devices specializing in nanoamps and picoamps, and for galvanometer circuits for detecting current nulls, but Fluke may be the first to put it in a standard bench multimeter. As an idea it is fine for measuring DC and low frequency AC currents. A lousy idea if you are measuring the current between a battery and a 1MHz switching power supply or any other ugly looking current. For most practical uses, you want a real resistor as the current sense resistor, and not a opamp input. If I had this Fluke feature in a bench multimeter, I would reserve the use for calibrating things, and use a handheld DMM or a current shunt for normal current measurements.

I don't believe IEC61010 requires the equipment to remain in spec or survive, just handle the transient without blowing up.

You are probably correct there. But I still think the Agilent designers would actually want their meter to survive if someone put 300V across the current range. Safety specs are only a bare minimum requirement, and Agilent see themselves as better then that. I suspect that if you look at the schematic of the current input, you will see that enough extra protection has been added so that it can survive. The lower the burden voltage, the harder it is to add comprehensive protection. They probably have something like a series fuse and and something like a thyristor connected across a bridge rectifier that is then across the shunt resistor. The Thyristor turns on when there is a current overload up till the time the fuse blows. This would limit the peak voltage across the current resistor to a few volts and so as long as the current resistor is no lower then about .5 ohms, no more then a few amps ever reach any precision part of the circuit. If you use the same circuit with a 0.01 ohm shunt resistor, you could get over 100Amps flowing through the current resistor before the fuse blows.

Richard

[alm]

There is zero error from the wiring and contact resistance. The current shunt resistor has taps for the voltmeter, and that bypasses all  wiring and contact resistance.

I'm not implying it causes an error, but that it's pointless to make the shunt resistor a much lower value than the other resistor, since the burden voltage is ultimately limited by the total resistance between the input connectors. Fuses also need to have a relatively high resistance since they need to dissipate enough energy to quickly vaporize the fuse wire in case of an overload.

As an idea it is fine for measuring DC and low frequency AC currents. A lousy idea if you are measuring the current between a battery and a 1MHz switching power supply or any other ugly looking current.

True. Since burden is a voltage of the opamps open loop gain, impedance/burden voltage is going to suck for high frequency signals.

For most practical uses, you want a real resistor as the current sense resistor, and not a opamp input. If I had this Fluke feature in a bench multimeter, I would reserve the use for calibrating things, and use a handheld DMM or a current shunt for normal current measurements.

The Fluke 8508A isn't exactly a practical everyday meter, it's a specialty reference multimeter that costs thousands of dollars. It is a neat trick for 'well-behaved' currents, though. I wouldn't mind seeing it becoming an option on more affordable meters.

I suspect that if you look at the schematic of the current input, you will see that enough extra protection has been added so that it can survive.

A quick check of the 34401A schematic (admittedly a lot older than the 34410A, but I don't believe its schematic is available) shows just two series fuses and a single 100 mOhm resistor to ground. The external fuse is just a glass F3A/250V fuse I think, the internal fuse is labeled as 7A, not sure about interrupt current rating. Funny how the burden voltage is specced < 2V for 3A, is the resistance in the fuses and contacts so large or is their spec extremely conservative? If there is 400 mOhm of series resistance in the rest of the circuit, maybe they just decided there was no point in reducing the shunt resistance further.

[amspire]

A quick check of the 34401A schematic (admittedly a lot older than the 34410A, but I don't believe its schematic is available) shows just two series fuses and a single 100 mOhm resistor to ground. The external fuse is just a glass F3A/250V fuse I think, the internal fuse is labeled as 7A, not sure about interrupt current rating. Funny how the burden voltage is specced < 2V for 3A, is the resistance in the fuses and contacts so large or is their spec extremely conservative? If there is 400 mOhm of series resistance in the rest of the circuit, maybe they just decided there was no point in reducing the shunt resistance further.

That actually sounds like a pretty lame attempt at protection for the current sensing resistors to me. I will see if I can find the circuit and take a look. Of course they might be so concerned about errors caused by leakage currents down at the uA level that they want to keep it very minimal.

[alm]

A second look shows a 4A bridge rectifier to ground that I missed. This should provide some protection, though I'm not sure how much help it's going to be at 300V or 1000V. With a 100 mOhm shunt resistor plus some series resistance, this should trip at around the same current as the 7A internal fuse.

[amspire]

I had a look at the 34401A circuit and it is a 0.1 ohm resistor, three relay contacts and a pair of fuses.

Given that the sense resistor has 0.3volts across it at 3A, and given that a normal 3A and 7A fuses have resistances of about 0.06 ohms and 0.01 ohms or 0.21V at 3A for the two fuses, how on earth do they get to the "<2v" figure?

To get anywhere near the 2 volt burden voltage number, it would need  0.5 Volts drop per relay contact. Fluke is quoting 0.15V total, including any relay switching.

Has anyone actually measured the burden voltage on a 34401A meter to see if it is near 2 V , or is it less then 0.6 V typically at 3A?

Richard.

[amspire]

A second look shows a 4A bridge rectifier to ground that I missed. This should provide some protection, though I'm not sure how much help it's going to be at 300V or 1000V. With a 100 mOhm shunt resistor plus some series resistance, this should trip at around the same current as the 7A internal fuse.

It has a 1.1KV varistor before the fuses and then a 3A and a 7A fuse in series.  Plus the bridge rectifier wires so that it will have a 2 diode drop to ground.

So if there is an overload, it will take something like 30A to blow the 3A fuse in half a mains cycle (10ms). A lot of that 30A will flown through the 0.1 ohm sense resistor, so the diodes are really there for even higher currents, of for protecting the low current sense resistor.

When a fuse blows, and inductive spike on the input is then absorbed by the varistor at a bit over 1.1KV.

So it all makes sense, The bridge rectifier limits the worst case current to the sense resistor  to something it can obviously cope with.

I also miscounted the relay contacts - it is only one and not 3. The other two contacts were the front/rear switch.  All the fault current has to go through this switch, so hopefully it has heavy duty contacts.

[ejeffrey]

The reason is simple: agilent loves options.  They sell current shunt accessories for their meters.  If their meters had low burden voltage on the high current settings, there would be less market for the options.

[alm]

The HP 3457A (the predecessor to the 34401A) and the Keithley 2000 also have 100 mOhm shunt resistors for the 3A range, but a < 1V burden voltage spec. I just measured the Keithley 2000, 0.7V burden voltage at 3A. So the switches and fuses contribute a little over 100 mOhm. Another DMM with a 0.1 ohm shunt a 3A fuse and two relays between the current jacks quotes 0.6V at 2A, but 2V with the optional scanner card (which adds some wiring and two additional relays). Are the designers expecting the relay contact resistance to increase sharply as they age?

The answer to this question about the absence of a 100 mA AC range suggests that the number of shunt resistors was a cost consideration.

[Rufus]

Has anyone actually measured the burden voltage on a 34401A meter to see if it is near 2 V , or is it less then 0.6 V typically at 3A?

My 34401A measured 0.86v @2.01A  and it climbed slowly as something was warming up.

[Zero999]

My 34401A measured 0.86v @2.01A  and it climbed slowly as something was warming up.

That implies a poor temperature coefficient.

Assuming you measured a fairly stable constant current, did the reading also increase as it warmed up?

[Rufus]

My 34401A measured 0.86v @2.01A  and it climbed slowly as something was warming up.

That implies a poor temperature coefficient.

Assuming you measured a fairly stable constant current, did the reading also increase as it warmed up?

Didn't surprise me, copper for the internal wiring and a +ve temperature coefficient for a fuse is probably good not poor. There was 1.7W warming up something.

[amspire]

My 34401A measured 0.86v @2.01A  and it climbed slowly as something was warming up.

That implies a poor temperature coefficient.

Assuming you measured a fairly stable constant current, did the reading also increase as it warmed up?

No, this really makes sense. It has two fuses in series with the current sense resistor, and fuses have a high positive temperature coefficient.

Is this the real explanation for the strangely high burden voltage?  At 2 amps, the sense resistor and contact resistances account for about 0.4V drop, and Rufus measured 0.86. So something like 0.46 volts (0.23 ohms) is across the fuses. At 3A, the fuses get hotter still, so the fuse resistance goes up to 0.4 or 0.5 ohms in total.

So how do the meters with a lower burden voltage do it?  One way is to only have one fuse, and secondly, the fuse resistance falls of very quickly as you make the fuse bigger. The Agilent has a 3A and a 7A fuse. Typical cold resistances of a Bussman 3A fuse is 0.06 ohms and the 7A is 0.01 ohms, so just a bit over double the current is a 6 times decrease. So if you can use a 10A fuse, or even a 20A fuse, you can get a much lower burden voltage. The cost is fault currents well into the several hundred amp range if you do put the current meter across a HV supply.

Agilent with their approach are able to limit huge currents from the current sense resistor, and the relay contact to that resistor, to something like 30A for 10 ms. A meter like the Fluke with a low burden voltage and a 0.01 ohm sense resistor would have to design so the sense resistor and relay can temporarily take several hundred amps for 10 ms without any damage. Can they actually do it? Or is Fluke really saying that they are making a precision lab meter and not a general purpose meter, so don't put the current range across a HV supply. It could be that Fluke make it safe, but not damage proof.

Richard

[alm]

The Fluke has a 10A range, so they must be using at least a 10A fuse. I believe it's got higher overvoltage ratings than the Agilent bench meters, so I can't imagine the protection being less effective. Maybe they spent some extra $$$ on a larger shunt resistor and more heavy duty wiring, so it can take the higher short circuit current? It has to be designed for 10A continuous anyway. They manage this in handhelds rated up to CAT IV 600V, so why not in bench meters? Only difference is the required stability of the shunt resistor and the use of relays instead of a rotary switch. I don't see any reason why a bench meter couldn't be designed to the same safety standards, it's just not very common since bench meters tend to be used in labs, not by industrial electricians.

[robrenz]

8846A

Designed to comply with IEC 61010-1:2000-1, UL 61010-1A1, CAN/CSA-C22.2 No. 61010.1, CAT I 1000V/CAT II 600V

Tool-accessible 11 A/1000 V and 440 mA/1000 V fuses, limits of
400 mA continuous 550 mA for 2 minutes on, 1 minute off.

• The 400 mA input is protected by a fuse (F2) rated at 440 mA, 1000 V (fast blow),
10,000 A minimum breaking capacity (Fluke Part No. 943121).
• The 10 A input is protected by a fuse (F1) rated at 11 A, 1000 V (fast blow),
10,000 A minimum breaking capacity (Fluke Part No. 803293).

Input Protection....................................................... 1000 V all ranges
Overrange............................................................... 20 % on all ranges except 1000 V dc, 1000 V ac (8846A),
750 V ac (8845A), Diode, and 10 A ranges

[robrenz]

So I actualy measured the burden voltage on my 8846A 
1 A = .0385 V  spec is <.05 V so there is .0285 Ohms in the 11Amp fuse, etc. plus the .01 Ohm shunt
3 A = .1155 V  spec is <.15 V ditto
6 A = .231 V ditto
only have 6 Amp CC supply so I couldnt test @10 Amp