Why is it that the low-current range on an ammeter (200mA) is fused and not the 20A range?
I recently lost a multimeter because I connected the ammeter in parallel with a power source by mistake.
All the ammeters I have lack a fuse on the 20A line, even the supposedly resilient ones in my university labs.
As Mr. Lathrop says, a fuse add cost and extra voltage drop. An even bigger problem, however, is fuses are not very good at protecting circuitry. In order for 200mA fuse socket to be useful, the ammeter must be able to consistently survive a substantial over-current condition during the time it takes the fuse to blow. At the 200mA range, that's probably not too hard. At the 10A range, it's a lot harder. Unless the 10A current shunt is massively over-engineered, it's likely to be damaged by any overload conditions sufficient to blow a 10A fuse. Further, most meters are designed with a philosophy that taking inaccurate measurements is worse than refusing to take any.
I would guess that in practice most ammeters are internally fused well enough that if connected directly to residential power mains they will usually open-circuit before they catch fire. Further, I would not be surprised if the readout and voltage-measurement circuitry was sufficiently isolated from the unexpected condition that it would not be electrically damaged. On the other hand, unless the current shunt is physically isolated from everything else, the other circuitry might still get damaged by molten or vaporized material from the shunt. And of course, "usually" doesn't mean "reliably".
Good quality multimeters do have fuse and other overcurrent protection on all their current ranges.
For example, Fluke multimeters can normally survive being connected to 250 VAC across their current measuring ranges
F87-V
F77-IV
Note separate High Rupture Capacity (HRC) fuses for both 10A and 400 mA inputs.
Some cheap multimeters omit these protections as a cost-saving measure.
See Multimeters that do not appear to meet their safety specs.
A fuse can add some voltage drop at high currents, and certainly adds cost. Consider that the cost is not only the fuse itself, which is actually pretty cheap, but the holder, some way for the user to access it, etc. This is basically a cost/quality tradeoff.
Some ammeters do have fuses on the highest current settings.
It isn't cheap to add safe and effective fusing to a meter. Until you get above about US $100 for the retail cost of the meter, you simply cannot afford to do it correctly.
These are the attributes we require for proper fusing in a hand-held industrial-grade DMM:
You must use an expensive HRC fuse here, which costs US $6-10 each in quantity 1k. (High-end DMMs are not high-volume items.) The ratio of BOM cost to retail cost is a factor of 3-5 for industrial products like DMMs, so this part alone adds $20-30 to the retail cost of the meter, making this the single most expensive part when you set out to add safe and effective fusing to a DMM.
(This in a world where $10 DMMs get serious consideration for some applications.)
You cannot use the cheap sort of fuse you normally see here, for a bunch of reasons:
Cheap fuses won't contain a high-energy blast. It's no good for the fuse to save the meter from a 100A spike if in doing so it throws hot slag over the PCB, bridging distant points so that the meter releases its magic smoke right after you replace the failed fuse.
Few fused devices are held in the hand while the fuse is sitting there deciding whether to blow or not. When you hook your 10A handheld meter up to a device that is about to fail by throwing a 100A spike across your measurement points, you really really really need the fuse to fail quickly and safely.
Cheap fuses are intended for applications where the normal steady-state current is known up front, and so that what you need is just a bit of margin over that value, plus maybe a bit of surge capacity to allow for the power-up inrush current.
We don't use DMMs that way. Even when we use DMMs correctly, we are purposely hooking them up to unknown currents and voltage sources, because we are using them to learn these values. When we misapply DMMs, such as your case where you dropped a current meter — nearly a dead short — across a power supply, we need the meter to protect the user first, and itself secondarily.
Cheap fuses tend to be unreliable. They are, after all, designed to break.
There's an old joke in hardware design: the most expensive part on the PCB will fail first in order to protect the fuse. You need to use really good fuses to flip that joke around to get a truth you're willing to live with.
The clips you need for the HRC fuses aren't free. They cost about 30 cents in quantity per pair, which adds $2-3 to the retail cost of the meter if we add a second set.
Keep in mind that we're distinguishing ourselves from the low end here, where you have two meters hanging on a peg in the hardware store, one marked $9.95, and the other $11.95, and the first one outsells the second, because hey, two bucks!
An HRC fuse constrains the kinetic energy of a failure within the fuse body, but we must also constrain the electrical energy. The sort of meter that has an HRC fuse or two inside is rated for Category III or IV testing. This means transient voltages of thousands of volts and effectively unlimited transient currents.
("Unlimited," you ask? If you drop a 0 gauge coppper rod across a power pole transformer's terminals, the copper goes away. That's as close to "unlimited" as matters for our purposes here.)
The space needed to constrain these energies increases the meter's cost.
Granted, the additional PCB space and plastic are not high costs, but again, we're creating a distinction in the product space here, separating our new meter design from the $10-20 meter class. What we spend here for a bit of extra PCB area and enclosure space — both in terms of resin and the higher cost of the larger mold needed to shape it — amounts to a significant fraction of the total retail cost of a cheap meter.
We also want to cut some slots in the PCB to allow for blast shields extruded from the case wall. (See RedGrittyBrick's Fluke pics.) These blast shields act as a backup in case the HRC fuse fails or its contacts arc over. The cutouts also replace some of the FR4 PCB material with air, which is a much better insulator.
PCB cutouts aren't free, but the biggest piece of the cost comes when adding the first one, which we already need for other reasons: around the input jacks, between the input section and the measurement section, etc. There is only a small additional cost to adding a few more cuts.
These extra costs are small; tens of cents.
All told, proper fusing adds about US $30 to the retail cost of the meter, per range. The two-fuse setup shown in RedGrittyBrick's Fluke pics therefore accounts for about $60 of the retail cost.
This is why a $50 "industrial grade" meter doesn't have an HRC fuse on both current ranges.
