If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

# LM231: LM231N frequency accuracy

Part Number: LM231

HI   Experts:

My responsible customer is using the LM231N to convert the voltage to the frequency ,  there is the frequency tolerance in same batch of boards, means the consistency of frequency is not good . It is around the 10% error .  According to the formula below , they checked every component in the formula ,and found the tolerance of the capacitor is 10%.

So can we think the frequency tolerance is caused by the capacitor ? And other factor?

### 4 Replies

• Hello Jason,
Sorry about the late response as I did not see your post earlier.

I could not understand the formula you are referring to. I've also moved your post to the Precision Amplifier group for better support.

Regards,
Hooman
• Hello Jason,

The LM231 is a precision V-to-F converter that is inherently precise by design. Its frequency of operation is described by the formula:

fout = (VIN / 2.09 V) x (Rs/RL) x (1 / RtCt)

The fout frequency accuracy is a function of VIN, and the passive components Rs, RL, Rt and Ct. If VIN is a set precision reference level, then the frequency accuracy will be a function of the listed passive components. Often, their importance is neglected.

Resistance tolerance of 1% and even 0.1% are commonly available and easy to apply. Yet, with a 1% resistor tolerance and its error contribution can be significant depending where any inaccuracy it introduces falls in the equation. Capacitors are often an afterthought and a 5% or 10% capacitor is often applied. This can be the greatest source of error contributing to frequency inaccuracy.

If for example the LM231 datasheet Fig. 14 circuit components are used for Rs, RL, Rt and Ct, and Ct takes on values 10% high (11 nF), and -10% low (9 nF), the respective fout frequencies will be 11.3 kHz and 9.27 kHz, with VIN set to 10 V - an 18% change. Then, there are the issues of the capacitor temperature coefficient and other dielectric characteristics that may affect its value.

The datasheet is specific about the component qualities for both the resistors and capacitors, “For best results, all the components should be stable low-temperature-coefficient components, such as metal-film resistors. The capacitor should have low dielectric absorption; depending on the temperature characteristics desired, NPO ceramic, polystyrene, Teflon or polypropylene are best suited.” The LM231 should not utilize low-cost ceramic capacitors having X7R, Y5V, Z5U, etc. dielectrics if accurate performance is important. Check the BOM and review the quality of the passive components being used.

Regards, Thomas

PA – Linear Applications Engineering

• In reply to Thomas Kuehl:

Hi Tomas:

Thank you for your detail explanation.

The temperature- coefficient you mentioned is the capacitance variation only depends on the temperature. Customer concerned about the capacitance

tolerance of the capacitors , which is relative to the consistency.

Perhaps customer could choose the low-temperature-coefficient components , but the consistency of the rated capacitance of the capacitors they can not

control for the mass production , and it will affect the fout accuracy among the different boards.

• In reply to Jason Fan:

Hello Jason,

Indeed, the temperature coefficient of the capacitor is an important consideration; especially if the board containing the LM231 is subjected to wide temperature changes. As mentioned, the datasheet suggests NPO ceramic, polystyrene, Teflon and polypropylene capacitors all of  which are more temperature stable than most ceramic capacitors. The cost of these other types of capacitors types is higher, they are usually larger in size for equivalent capacitance, and the capacitance values may be more limited when compared to the low-cost ceramic capacitors. So it looks like it is a matter of identifying the best compromise between capacitor accuracy and stability, and cost.

Regards, Thomas

Precision Analog Applicaitons Engineering