Microelectronics | VFD Tube VU Meter

Synopsis

I have recently acquired a 2 x 14 element vacuum fluorescent display for use with VU meter circuits. The tube is of type ИЛТ3-15М (ILT3-15M) and was originally meant to be controlled by the КР1534ПП4 (KR1534PP4) integrated circuit. Since those ICs are both scarce and prone to failure, I had to design my own circuit. The tube assumes a multiplex driving technique, as there are two command grids and all fluorescent elements are connected in parallel. I have thought about a use for this VFD tube, and the best fit was a VU meter circuit for my Audio Sources Selector.

And since I like weird circuits, I thought about an appropriate schematic diagram. I plan to use a CD4016 integrated circuit to alternatively switch between the LEFT ad RIGHT audio channels, also the command grid voltages. Does this classify as non-standard use? As for the oscillator, I was thinking about a CD4047 dual-mode multivibrator.

Now regarding the actual VU meter circuit, I went with a standard implementation based on the LM324 circuit. In fact, I used four of them for those fourteen fluorescent elements. And I used the remaining two operational amplifiers as input signal buffers.

Schematic Diagram

For starters, let's take a look at a block diagram as it depicts the principle of operation. As said above, there are two command grids that must be activated alternatively, one for each audio channel. A small audio buffer is required to separate the audio inputs from the actual VU meter circuit. And finally, the driver block will command the lighting of the individual fluorescent elements. One caveat, though, is that I cannot calibrate the outputs of the operational amplifiers to the exact dB readings displayed by the tube.

Now let's take a look at the electrical schematic diagram.

Fig. 1: Electrical Schematic Diagram

Each of the fourteen operational amplifier used to drive the individual fluorescent elements (cathodes) are connected as comparators. The time-multiplexing technique I used applies alternately the LEFT and RIGHT audio signals to both VU display units at once. But since the grids are also time-multiplexed, only one channel will be visible at a time.

CD4016 is connected as a dual 2:1 multiplexer, and CD4047 is wired as variable frequency oscillator. The multiplexing frequency has to be carefully adjusted so that there is no eye strain caused by flicker due to rapid changing of command grid voltages.

Construction and Pictures

Here are some pictures of the parts needed for this VU meter.

Initially, I went with air wiring to test whether my approach is worth implementing. The first test used only one LM324 to light the first four fluorescent display elements. The range displayed by the tube is -20 dB to -8 dB. But now we know that it is incorrect.

In addition, my tests revealed that the grids should be more positive than each individual element control voltages. Unfortunately the tube lights up pretty dimly since the electrical potential difference between the grid and the cathodes (driven by outputs of LM324) is very small. In practice, the grids should be powered at a few tens of volts higher than the cathodes. I'd say a 25 V to 35 V would be sufficient for the grids. But in this case, CMOS power switching is out of the question. And electromechanical switching is virtually impossible. A better solution would be to use individual transistor drivers, controlled by CD4016.

The printed circuit board is dual-layer. I've printed them on translucent film and on paper as a template for the manually drilled holes. The red plastic filter is intended to be mounted in front of the VFD tube.

The integrated circuits were made by Microelectronica back in the 1990s. These are standard CMOS chips, directly equivalent to CD4066 and CD4017. Since the original labels have faded away, I've glued printed labels for easier identification. That's not really proficient, but for testing purposes, anything is acceptable.

At this point, I am convinced the circuit in its current implementation is just another proof of concept. For sure, the control grid power switching has to be done with individual transistor drivers. And the outputs of the individual operational amplifiers can be calibrated to the values expected by the VFD tube. For the record, the values are: -20 dB, -15 dB, -10 dB, -8 dB, -6 dB, -4 dB, -3 dB, -2 dB, -1 dB, 0 dB, +1 dB, +2 dB, +3 dB, and +6 dB.

In any case, you can use this design as a starting point for something better.

These kinds of projects don't always succeed from the first try. But they are a knowledge gold mine. At least, I've learned a lot about multiplexed VFD VU meters. However, the visual reward of the green and red phosphorus glow is worth the effort. And since the ИЛТ3-15М tube is hard to find, the projects simply become more exotic.

For sure, this kind of VFD-based VU meter, once correctly implemented, would be a great addition to any DIY audio gear.

Thank you for reading this page, and I hope you found it useful.