EEVblog Electronics Community Forum

Electronics => Projects, Designs, and Technical Stuff => Topic started by: 001 on November 30, 2016, 01:17:18 pm

Title: few MHz true RMS detector project (schematic and pcb inside)
Post by: 001 on November 30, 2016, 01:17:18 pm
Hi!
I`m looking for ideas for ~5MHz true RMS detector schematic (AD637 avoided since it only 10kHz, thermoresistors avoided too)

The one project I grab in attachment.  (is it ok? what nowday opams and 'few-transistor-on-one-crystal' IC I can use?) Please comment.

google translate https://translate.google.ru/translate?sl=auto&tl=en&js=y&prev=_t&hl=ru&ie=UTF-8&u=http%3A%2F%2Fwww.irls.narod.ru%2Fizm%2Fvolt%2Fskv.htm&edit-text= (https://translate.google.ru/translate?sl=auto&tl=en&js=y&prev=_t&hl=ru&ie=UTF-8&u=http%3A%2F%2Fwww.irls.narod.ru%2Fizm%2Fvolt%2Fskv.htm&edit-text=)

full article with PCB http://irls.narod.ru/izm/volt/skv.htm (http://irls.narod.ru/izm/volt/skv.htm)
Title: Re: few MHz true RMS detector project (schematic and pcb inside)
Post by: 2N3055 on November 30, 2016, 04:57:10 pm
Linear technologies LTC1968 goes up to 15MHz -3dB, to 500kHz -1%,  Works really well up to 150kHz with 0,1% error..

If that is enough bandwidth, very easy to apply..
Title: Re: few MHz true RMS detector project (schematic and pcb inside)
Post by: Kleinstein on November 30, 2016, 06:22:47 pm
I don't think it is realistic and easy to get faster with a discrete RMS converter.

Besides the analog way, there is also the option to digitize the signal and do RMS in software, like some Keysight DMMs do ("True Volt"). Today a µC like an STM32F3... with an 5 MHz range 12 Bit ADC could do that quite fast. It is also quite accurate. Due to massive oversampling the resolution is better than the 12 Bits of the ADC.
Title: Re: few MHz true RMS detector project (schematic and pcb inside)
Post by: 001 on November 30, 2016, 08:09:28 pm
Thanx for Yours answers!

Digitalisation and absolute tolerance is not nesessary for me since I use millivoltmeter to compare levels in dB

Any ideas for this analog detector?
Title: Re: few MHz true RMS detector project (schematic and pcb inside)
Post by: mk_ on November 30, 2016, 08:41:34 pm
Quote from: 001 on November 30, 2016, 08:09:28 pm
Thanx for Yours answers!

Digitalisation and absolute tolerance is not nesessary for me since I use millivoltmeter to compare levels in dB

Any ideas for this analog detector?

AD8307... and some tricky (aka black magic) analog frontend could do what you asking for if "db" is enough.


Michael
Title: Re: few MHz true RMS detector project (schematic and pcb inside)
Post by: MatteoX on December 01, 2016, 06:11:01 am
If you just want to have a true RMS voltmeter, maybe you should consider HP 3400A. It can be obtained on ebay, with some patience, very cheap. Probably cheaper than something comparable you can build.

If you want to experiment, you coud try to use square law diode detector.

Square law diode detector is used in, for example, Ballantine 323 analog true rms voltmeter. This meter from early 1970s still has quite respectable specs (10Hz - 20 MHz,  the lowest range is 100uV full scale). Schematics is available online. After squaring and rectifiying, the signal (I believe less than 500uV max) is chopped and amplified. It has electromechanical chopper operating at 94 Hz.

Ballantine still makes this meter. I guess it does not have electromechanical chopper anymore, but I couldn't find the schematics for the current model (they sell manuals for $100!!!)

I got one recently very cheap but the mechanical chopper is not working. When I got time I'll try to replace it with an electronic one.

Title: Re: few MHz true RMS detector project (schematic and pcb inside)
Post by: Vgkid on December 01, 2016, 06:22:22 am
From what I can remember. Schematic/parts wise the 323-07 is about the same as the older 323. On my -07 the power supply is made from regular voltage regulators ex. 78xx series.