Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

best part for current sensing

Status
Not open for further replies.

carmusic

New Member
Hi i need the best way to measure small current from a power supplky, it passes through a 0.01 ohm shunt at positive terminal, i was using tle2071 op amp for that in a differential connection but i always get offset (+/- 100mv) since my resistor are not precise enough (i use +/-1%) to get a good precision at high gain (i need about 180 gain)

is there ICs that we connect directly to the shunt and output a high voltage (i need about 0-10V for 0-50mv signal)
 
Everyone, including me, assumes that the classic bridge circuit is the best way, but I found this circuit from National Semi many years ago and it is far better as it has none of the resistor sensitivities that the bridge circuit has plus offsets are easily tuned out. I've used this in many products and it works very well.

You must use a rail-to-rail input op amp since the power supply rail is virtually at the same voltage as the inverting input and it is best if the output is rail-to-rail too. I'm sure this circuit will work at lower voltages than 12V as long as the FET has an appropriate Vgs(th) somewhere at or below mid Vcc. The 2n7002 is a good general fit for this.

Rs is the sense resistor (like .1 ohms or such) and R1 and R2 are usually about 1K and 10K for example. For measuring very small currents, consider a larger sense resistor and increase R2.
 

Attachments

  • current sensor 2.jpg
    current sensor 2.jpg
    162.8 KB · Views: 532
Last edited:
Ron, most or all of the parts point to above use essentially that same circuit, but they are integrated into the part (except maybe the sense resistor).
 
Roff said:
Ron, most or all of the parts point to above use essentially that same circuit, but they are integrated into the part (except maybe the sense resistor).

Oh. OK. Not surprised, I guess. Bit embarrassed maybe. Tempted to say "never mind" like we used to see on Saturday Night Live. Are those parts cheaper than the discrete approach?
 
RadioRon said:
Oh. OK. Not surprised, I guess. Bit embarrassed maybe. Tempted to say "never mind" like we used to see on Saturday Night Live. Are those parts cheaper than the discrete approach?
I just did a quick check on MAX4376 (this is not a recommendation). Digikey has them for US $0.85 in unit quantity.
I haven't used any of these parts, I just am aware of them.
 
Roff said:
I just did a quick check on MAX4376 (this is not a recommendation). Digikey has them for US $0.85 in unit quantity.
I haven't used any of these parts, I just am aware of them.

That's not too bad. In my most recent employment at a manufacturing company, our operations group strongly discouraged us from designing-in Maxim parts due to their claims of supplier "inflexibility". But I realize that this part is only one example of many manufacturers.
 
RadioRon said:
That's not too bad. In my most recent employment at a manufacturing company, our operations group strongly discouraged us from designing-in Maxim parts due to their claims of supplier "inflexibility". But I realize that this part is only one example of many manufacturers.

Going back years now, Grundig still used discrete IR receivers in their TV's, when everyone else had gone to the small three pin IC's - this was because Grundig wouldn't use parts they couldn't source from more than a single manufacturer.

Perhaps it's the same at your employers?.
 
Nigel Goodwin said:
Going back years now, Grundig still used discrete IR receivers in their TV's, when everyone else had gone to the small three pin IC's - this was because Grundig wouldn't use parts they couldn't source from more than a single manufacturer.

Perhaps it's the same at your employers?.
When I worked for a large DRAM manufacturer, I was involved in many meetings with our competitors (4 or 5 of them) so that we made certain that the next generation of DRAM would work the same, no matter whom you bought it from. Single-source commodity parts are not commodity parts.:D The competition came when we tried to be the first with the next speed improvement, at the lowest cost.
 
I can see their point, avoiding vendor lock in can often be a good idea. However, if the part can be easily subsituted with minimal re-design and isn't the main part of the design, it doesn't matter.
 
Nigel Goodwin said:
Going back years now, Grundig still used discrete IR receivers in their TV's, when everyone else had gone to the small three pin IC's - this was because Grundig wouldn't use parts they couldn't source from more than a single manufacturer.

Perhaps it's the same at your employers?.

No, it wasn't quite as bad as that. The way I heard it, our sales were growing fairly erratically and the factory was faced with sometimes asking for delays in material delivery, sometimes speedup of delivery, sometimes returns were necessary, sometimes they needed to change quantities in an order, perhaps there was some requests for unusual credit terms, I don't know. But these are typically the things that the factory buyers needed flexibility on from part vendors, and Maxim apparently just said NO to every request for every adjustment to delivery dates, quantities on order, or whatever. This was rare as virtually all other vendors were negotiable on these sorts of things.
 
Hero999 said:
I can see their point, avoiding vendor lock in can often be a good idea. However, if the part can be easily subsituted with minimal re-design and isn't the main part of the design, it doesn't matter.

In RF work, single sourced parts, especially for the ICs and filters, was pretty normal and made it harder for the factory buyers.

Even minor redesigns could affect our carrier certifications and regulatory approvals so we didn't do minor adjustments to hardware very often (as opposed to firmware which got "fixed" very often). Did you know that while it might cost, say for example, $2M in labor and direct expenses to design a cellphone and produce some prototypes, it might cost more than $800,000 just to do all the certifications and approvals for carriers (the system operators like Verizon etc.) and governments so that you could sell it. So, having to redo these as a result of hardware changes was not popular.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top