For decades, I’ve had a fascination with voltage-to-frequency converters and the 555 analog timer chip, and therefore a double obsession with VFCs based on the 555. In fact, my first Design Ideas (DI) submission (in 1974) was for a 555 VFC. It was not only published but also selected as the best DI of the year. That was it, I was thenceforth hooked forever.
Wow the engineering world with your unique design: Design Ideas Submission Guide
The simple 555 VFC design to be presented here, so far as I know and as unlikely as it sounds for so *ahem* “mature” a part as the 555, is novel. It runs with good linearity and accuracy at 1 MHz, with even faster operation possible. That’s 100x faster than that 1974 555 frequency converter.
I hope you’ll find its details interesting. Here’s how it works. The story begins with Figure 1.
Figure 1 Starred components are precision, including the +5 V supply, but something’s missing.
There’s nothing novel about the input current source comprising A1, Q1, and surrounding parts. It supplies 0 to 1 mA to the U1 current-to-frequency converter in response to its input voltage, as scaled and offset by R1 and R2. The values shown set a 0 to +5 V input span. R1 = 1.8M and R2 = 200k would make it -5 V to +5 V.
A capacitor added in parallel with R2 will provide extra noise rejection. But the inherent noise immunity of the VFC analog-to-digital conversion is good, so you probably won’t need it.
Moving further into the circuit is when things do start to get weird, because the usual two resistors associated with 555 oscillators are missing. Also missing is the usual astable 555 1/3V+ peak-to-peak voltage swing. This topology generates a 2/3V+ Vpp linear sawtooth waveform that resets, not to V+/3, but to zero. Unfortunately, while the sawtooth is nicely linear, due to U1’s internal switching delays Td, the frequency versus Q1 current Iq1 relationship is not very linear. Figure 2 shows how bad it is:
Frequency of oscillation (FOO) = 1.0/((VthC2/Ir3 + Td) = 1.0/(1.0ns/Ir3 + Td)
Figure 2 Nonlinear red curve versus ideal black shows ~20% linearity error from LMC555 internal delays.
Luckily, as derived in another recent DI: “Improve 555 frequency linearity.“
…it’s an easy fix. It consists of a single resistor, R4, connected between the Dch (discharge) and Thr (threshold) pins. R4 is used to linearize the current-versus-frequency function by biasing the Thr pin upward by IcR4. That cuts short the duration of the positive-going timing ramp and thereby the sawtooth period by the same amount that the delays lengthen it: IcR4/(Ic/C2) = R4C2 = Td.
Thus, if R4 is chosen so R4C2 = Td as shown in Figure 3, nonlinearity compensation will be (at least theoretically) complete over the full range of control current. The frequency of oscillator (FOO) for this circuit:
FOO = 1/((VthC2)/Iq1 + 212ns – Td) = 1/(1.0ncb/Iq1 + 212ns – 212ns) = 1/(1.0ncb /Iq1) = 1000 Iq1 MHz = 1MHz(+5v – Vin)/+5v
Figure 3 R4C2 = Td = 212ns = nonlinearity compensation for 555 internal delays.
Now FOO will linearly track Iq1 and therefore Vin as shown in Figure 4.
Figure 4 Nonlinearity disappears if R4 = Td/C2 =212 ns/300 pF = 706 ohms.
Stephen Woodward‘s relationship with EDN’s DI column goes back quite a long way. Over 200 submissions have been accepted since his first contribution back in 1974. They have included the best Design Idea of the year in 1974 and 2001.
Related Content