# LED ammeter using lm3914

Status
Not open for further replies.

#### johnsenior

##### New Member
I want to build a LED ammeter to supplement the existing analog (needle style) ammeter in the power supply for an Electric Vehicle. I plan to use the typical voltmeter circuit based on a lm3914 ten segment LED driver, and just tap into the sensing leads to the existing ammeter. The 12V auxiliary circuit will be used to power the lm3914. (The electric motor runs on 144V from 24 six volt batteries).

However, the shunt built into the EV power supply to feed the ammeter only provides 50 mV drop at full power (50 amps). I assume this is less than is needed to drive the lm3914 to full scale. Can anybody suggest a simple addition to the circuit to boost the voltage enough to use the lm3914 IC? Or is there a better approach?

#### audioguru

##### Well-Known Member
An LM3914 needs a minimum of 1.25V for full scale. An opamp can amplify your 50mV to 1.25V. A few opamps do not need a negative supply but they all need a positive supply.

#### kchriste

##### New Member
Forum Supporter
I assume the shunt is in the 144Vdc positive lead of the battery string. Be very carefull with this! If your 12V auxiliary supply is not isolated from the 144Vdc, then expect a very large arc and smoke if you connect them together (even just one lead)!
Most current-shunt monitor IC's won't go that high (144Vdc) so you might want to look into low side monitoring. This IC may work, but high frequency noise on the 144Vdc line will be a problem for it.

Last edited:

#### MikeMl

##### Well-Known Member
Follow this link to the data sheet. Look at the application note. By picking one resistor, you can convert the 50mV differential voltage across your shunt to a ground referenced voltage to drive your LED bar graph chip.

Last edited:

#### kchriste

##### New Member
Forum Supporter
Unless I'm missing something, the ZXCT1009 is limited to a max voltage of 20V. Johnsenior needs at least 144vdc if sensing the high side.
A low side shunt, if possible, is probably the best solution.

#### MikeMl

##### Well-Known Member
You are right. I missed the 144V high-side voltage.

#### tcmtech

##### Banned
What I have seen of electric vehicles the amp meter shunts are on the negative side. This keeps the voltage across the meter leads in closer reference to the body. (assuming it uses a negative grounded or referenced drive system.)
The whole drive power circuit could be completely floating too!

just a guess!
Although a ten segment readout is less accurate than a mechanical meter.
An ICL7107 3.5 digit LED digital meter IC would give you actual numbers!
And digital amp meters based on this IC are readily available on EBay for around $10 shipping included. I just purchased a few more a week ago. The volt meters and amp meter circuits are very easy to hack into and Change the actual readout rates and input ranges. Plus the IC is socket mounted so it you fry it (I did one all ready!) you just pop it out and snap a new one in and your good to go! A replacement IC is about$3 on eBay too!
YOU cant beat the \$10 price either for a complete ready to run digital meter!

#### johnsenior

##### New Member
Thanks all for the great comments. I think I have enough to proceed now. It has taken a while for me to get up to speed. (some senior moments are longer than others).The shunt is on the low side. However the 12v auxiliary battery is charged from the 144v system, probably with a dc/dc converter. I will check to make sure there is no 144V leakage. If there is I will use an independent battery to supply the ammeter. I plan to use an op-amp to boost the signal input to the lm3914.

I have seen the low cost digital LED amp meters and plan to pick one up to try. However, my unstated objective in building the ammeter is to help the driver avoid her lead-foot driving that shortens battery life. I think a flashing red LED will be better at getting the drivers attention than an ammeter reading of 33.2 amps. Thanks again for the input. Any other thoughts would be appreciated.

Status
Not open for further replies.