# N-MOS real-life gate threshold voltage (Vgsth) lower than claimed

I faced with a paradox that N-MOS transistor real-life Vgsth (gate-to-source threshold voltage) is significantly lower than claimed. It should be about 1.7 V (and the simulation confirming that assumption):

However, real-life test shows that the Vout voltage is about 0.6V (real-life test schematic was similar).

During further investigation I was lucky to find a "Sub-threshold drain current" chart in NXPs datasheet page 7 (usually there is no such data in MOSFET datasheets) for this MOSFET (2N7002):

As you can see, at my given current (0.5 mA) Vgs should be miminum above 1V (I'd say 1.1V).

Is there anything I left outside my sight?

Actually the main reason I am asking this question is that I need this schematic to provide me with Vout as low as possible (0.6V is not a bad result, but the closer Vout to 0 V the better). But this is a bit scary to put in mass production a schematic which not behaved in accordance with data sheet. So I'd like to pick a different MOSFET with as low Vgsth as possible.

• Your circuit cannot deliver 0.5 mA. Clearly 26 volts/68 k is 0.38 mA. Where did you get the device from? – Andy aka Aug 11 '16 at 8:19
• @Andyaka, sorry for this mistake: I tested this schematic up to 51V with the same result. I made needed changes to original question – Roman Matveev Aug 11 '16 at 8:24
• There are a lot of "fake" stuff around. – Andy aka Aug 11 '16 at 11:01
• Worth also checking several samples : marginal ESD damage increasing leakage current could deliver the same result, and you'd never know it was damaged if you only tested one. – Brian Drummond Aug 11 '16 at 17:02
• That subvt graph is sort of freaky. Basically, it shows that you Vth variation of 1.3 volts, but considering that you get 1 decade of current for 300mV (the theoretical maximum is ~70mV), I would say that you do indeed have a device that just happens to have a lower threshold. The 1.3 spread on subvt suggests that the parts are run on several fabs, and that means that you will indeed have a lot of variance. – b degnan Aug 11 '16 at 17:15

There is no problem here, you can relax. The mosfet is behaving exactly like it should! If you try a dozen of them, you'll get the same result. And that result is exactly what is claimed. In fact, it's just what NXP says it should be.

This 'paradox' is actually quite easily explained.

You're attempting to use a crappy level 1 mosfet spice model from 1973 to simulate sub-threshold effects. I know you're using a crappy level 1 model from 1973 because that's the only kind of mosfet models LTspice comes with. They are useful for simulating power properties of power mosfets, and they're intentionally simplistic to speed up simulations. Which is great, LTspice is unmatched at DC/DC simulation speed.

What those models can't do is pretty much anything else. They won't even accurately model threshold effects, and forget subthreshold effects.

SPICE is only as good as the models you use. MOSFETS are particularly tricky, and as our lithography processes have gotten finer, so has our understanding of MOSFETs. They're very complicated beasts with all sorts of quantized phenomena to take into account. Indeed, the specific property you're trying to exploit, sub-threshold current, is dependent on the fact that electrons are quantized (so, it involves quantum effects) just for starters.

Then there are a multitude of short channel effects that definitely effect something like the 2N7002, where the channel length and depletion layers are the same order of magnitude in size. The reason you can't find these graphs in datasheets is because they aren't useful. That graph in particular doesn't even tell you what the Vds was when they measured it. Considering that the sub-threshold current has an exponential relationship with that value, its omission renders the graph worthless. Sometimes NXP will include it, but it will make the chart useful ONLY if all your conditions are likewise identical.

Certain things are just too complex to accurately convey using one chart in a datasheet. This is one of them. A sure sign is if its unusually difficult to find this information in datasheets. It's not just what datasheets tell you, but also what they don't tell you that can be important. Often, the not-telling is as intentional as the telling.

I actually have chosen MOSFETs entirely by how good their SPICE models are. For example, Fairchild actually has very good models for some of their power mosfets. Nice Level 7 models. On Semiconductor, on the other hand, has always had really crappy models, level 2 or 3. Yes, the irony that they're now the same company is not lost on me. But I digress. If you're going to use SPICE in any form for real production work, you can't just hope for the best when it comes to your models. Open them, look at them, learn how to read subcircuits, and know the model you're using is good enough for what you need.

A quick and dirty estimate however is just to look at the model number. Higher is better. Here is a more depth description.

Level 7 is sort of the minimum I would recommend for most stuff, but something fairly simple in construction like 2N7002, you can actually get quite good subthreshold modeling with a level 3 model.

Anyway, I grabbed the more than adequate model from NXP for their 2N7002, and pasted it directly into LTspice (this way, the model is part of the .asc schematic capture file and instantly portable, and you can easily inspect the model most important part of this circuit - the mosfet), set the prefix of the 2N7002 to XN instead of MN (X tells LTspice to look for a proper subcircuit model, instead of using it's crappy 1973 models), and that's all there is to it.

Want to take a guess at what a proper model thinks your Vout should be?

That's right. It thinks your Vout should be exactly what it really is.

SPICE is awesome. Just make sure your models are too!

• Nice post, one thing though: the prefix for subcircuit should be X, not followed by any other letter. Only transistors have QN/QP and MN/MP, to distinguish them in terms of symbols, though they are, in fact, one an only .model, internally, the only difference being an attribute. – a concerned citizen Oct 3 '16 at 15:20