In the Facebook etracer group a member asked how important it is to supply a test voltage beyond 400V. Why can't we just test a tube up to 400V and extrapolate the curves from there. Well, besides the trivial answer that extrapolation is not reliable there are actually situations where extrapolation simply doesn't work.

I am not sure since when vacuum tube testers like etracer are categorized as "pulse-type" tube testers. And this type of tube testers are usually being criticized as not accurate because the DUTs are not biased at the quiescent point and hence during testing the temperature of the DUT is lower than the temperature of the DUT in the real circuits. However, little information can be found on the accuracy requirement. How bad is it? And how much deviation is acceptable? 1%? 10%. In this article I will explore this issue a little deeper and measurement data are provided for references.

The etracer software GUI


The GUI (Graphics User Interface) in the etracer software is simple yet effective. It provides two modes of measurements: Quick-Scan mode and Full-Scan mode.

In the Quick-Scan mode the user specify the quiescent operation point for a vacuum tube DUT (Device Under Test) and the software output the measured parameters at that operation point. The Output parameters are: mu, gm and rp. The following is an example output of a measurement made on a 12AX7. Both section are measured simultaneously.

Example output of the Quick-Scan measurement of a 12AX7


In the Full-scan mode the user defines the sweeping ranges and sources for the plate electrode, the grid electrode and possibly a second grid electrode (For pentodes) or a second plate electrode (For Twin-triodes). The software performs a quick curve-tracing following the configured parameters and the curves will be spline-interpolated at the end of curve-tracing to align the data points to the grid on the X-axis. The data points can be saved to a file in the CSV (Column Separated Values) format for further analysis such as matching. The following are some example outputs for a full-scan. The following are a few examples of full-scan measurements.


As I mentioned in one of my earlier article the initial goal of etracer was an enhanced utracer. The project turned out to be a totally different design and shares nothing in common except the idea of using a capacitor to supply high voltages. the letter 'e' in etracer thus has multiple meanings. It can be enhanced, electron or essues. The key differences between etracer and utracer are listed below:

  etracer utracer
High voltage supply range 0~750V 0~400V
Negative supply range 0~-170V 0~-40V
Heater supply 1.5V~27V regulated DC supply with a current capacity of at least 3A. One side of the heater supply is always connected to the system ground by a 0.1ohms current sensing resistor. Heater voltage is emulated by chopping the 19V DC input. Or an external DC supply is required.
High voltage supply topology Output voltage is 0~750Vdc referred to the system ground. Output voltage is 19V above the system ground. When making a measurement the heater needs to be disconnected from the system.
Voltage charging time Sophisticated DSP algorithms ensuring accuracy in the low-voltage range and fast charging time in the high voltage region Fixed charging period. Charging time is significantly longer in the high voltage range (>200V)
ADC resolution and sampling rate 14 bits/ 900k samples/sec  10 bits

Beside the technical differences etracer is offered as a built PCB while utracer is in un-built kit format. etracer also have a companion chassis, the Model-01 to save the users' precious time and possible frustration.

This site is dedicated to the development of etracer, the 'enhanced' vacuum tube curve tracer. This project was inspired by the utracer designed by Ronald Dekker (DOS4EVER). I purchased a utracer kit from Ronald in late 2015, assembled it and started to use it soon after. After using utracer for like two to three times I figured there are many things lacking in it. As a DHT (Directly Heated Triode) lover a lot portion of my vacuum tubes collections are power triodes like 2A3s, 300Bs and 50s. These DHTs need much more than -40V  on the grid to give a complete curve scan. The maximum anode voltage of 400V on utracer seems to be adequate but for a complete curve scan I would like to go above 600V. Also the filament supply on utracer supplying 'equivalent power' by chopping a 19V DC supply is a little bit scary when testing valuable tubes. Nonetheless, Ronald's website has a lot of valuable information on the design of utracer. I strongly recommend readers to visit his utracer website:

I really like the nice and clean design of utracer hence I offered Ronald Dekker a proposal to collaborate and improve the utracer based on what Ronald's design. Unfortunately Ronald didn't like my proposal and hence I decided to start from scratch.

The etracer project was kicked off in April 2017 and I built the first prototpye on a perfboard. The main goal of the prototype was a viability study going above 400V. Through intensive studies on the internet I figured a way to bring the test voltage well above 1kV. However, due to the limited components selection and real estate concern I decided to stop at 750V. For the negative supply I adopted the same capacitor-charging principle and made it adjustable from 0V to -160V. I also designed a simple DC-DC converter to supply a maximum of 3A current from 1.25V to 26.5V for the filament.

The first etracer on a perfboard

 With the success of the perboard, I moved on and laid out a PCB using the open source software 'KiCad' and sent the Gerber files to a PCB manufacturer in China. The first PCB of etracer was received in June 2017 and was stuffed and verified within a week. The resulting product has an astonishing high voltage specification, a footprint of merely 21cm by 15cm. It is very power efficient due to the fact that it doesn't deliver current to the DUT (Device Under Test) during idle time. It can be powered by a DC source of 29V with a current specification over 3A. The components on the PCB draw less than 2 Watts during idle time. When the heater supply is turned on it will deliver power at specified voltage to the filament of the DUT with an conversion efficiency of at least 85%.  When measurement starts the board will consume a maximum of 50 Watts more to charge the capacitors during a very short period of time.

Fully assembled etracer PCB V1.1