Hi everyone.
I recently re-calibrated my old signal generator (Farnell LFM 4).
Most things went well. I realised that there was a problem with the square signal (trim pot couldn’t go low enough for calibration - it was asking 14v p-p = meaning 4.92 rms and the minimum i could take it was 5.35 rms) but i don’t ever use square signal, so all ok.
The all important sine wave calibration went well. But I realized that there is a slight problem with the lowest scale.
The scales are from about 0.5v - about 14v, then from about 50mv- about 1.4v, the third from about 5mv - about 140mv, and the lowest from about 0.5mv - about 14mv. All p-p.
While on the 10mv scale (5mv-140mv p-p) i can go as low as 4mv p-p abd measure 1.3mv rms
But on the lowest scale of 1mv (0.5mv- 14mv p-p) the readings up to 5mv are a bit higher than the low settings of the previous scale. I accept the fact that the lowest scale may be more accurate on lower voltages and gives more accurate readings.
But even the lowest (near zero reading of that scale ) which lasts from 0mv to 4mv (maybe a meter or cable problem to detect readings in lower settings) is higher than the readings of the previous scale. The 10mv p-p scale gives initial readings of 1.3mv rms, the 1mv p-p gives initial readings of 1.6mv rms.
I have to admit that i can work with these settings. But if i could improve something easily why not?
The unit is super clean . Had never been opened (mine was the first ever attempted calibration on the unit - there were seals to prove it all over it). No switch had any obvious problem- no bad cap in sight.
So what can cause such an issue . Maybe a general “inherited” issue of the device because of construction and age of components ?
All measurements are taken with a 2 seperate banana plug cables (not a double banana plug to a ciaxial cable) from the signal generator to pl 259 on the meter.
Maybe straight coaxial to bnc - banana adapter (instead of 2 banana cables soldered on coaxial ) on the signal generator would help to reduce rf and noise?
I recently re-calibrated my old signal generator (Farnell LFM 4).
Most things went well. I realised that there was a problem with the square signal (trim pot couldn’t go low enough for calibration - it was asking 14v p-p = meaning 4.92 rms and the minimum i could take it was 5.35 rms) but i don’t ever use square signal, so all ok.
The all important sine wave calibration went well. But I realized that there is a slight problem with the lowest scale.
The scales are from about 0.5v - about 14v, then from about 50mv- about 1.4v, the third from about 5mv - about 140mv, and the lowest from about 0.5mv - about 14mv. All p-p.
While on the 10mv scale (5mv-140mv p-p) i can go as low as 4mv p-p abd measure 1.3mv rms
But on the lowest scale of 1mv (0.5mv- 14mv p-p) the readings up to 5mv are a bit higher than the low settings of the previous scale. I accept the fact that the lowest scale may be more accurate on lower voltages and gives more accurate readings.
But even the lowest (near zero reading of that scale ) which lasts from 0mv to 4mv (maybe a meter or cable problem to detect readings in lower settings) is higher than the readings of the previous scale. The 10mv p-p scale gives initial readings of 1.3mv rms, the 1mv p-p gives initial readings of 1.6mv rms.
I have to admit that i can work with these settings. But if i could improve something easily why not?
The unit is super clean . Had never been opened (mine was the first ever attempted calibration on the unit - there were seals to prove it all over it). No switch had any obvious problem- no bad cap in sight.
So what can cause such an issue . Maybe a general “inherited” issue of the device because of construction and age of components ?
All measurements are taken with a 2 seperate banana plug cables (not a double banana plug to a ciaxial cable) from the signal generator to pl 259 on the meter.
Maybe straight coaxial to bnc - banana adapter (instead of 2 banana cables soldered on coaxial ) on the signal generator would help to reduce rf and noise?
Last edited:
"14v p-p = meaning 4.92 (v) rms" of "the square signal" is not true.
For a symmetrical, bipolar square it is 14 ..
For a symmetrical, bipolar square it is 14 ..
Thanks for that. Good to know. I know nothing about square. I just used the sine wave formula even on square.
Does it influence the sine wave scales (from the schematics)?
Does it influence the sine wave scales (from the schematics)?
Yes, the RMS values are different for sine and square waves of the same peak amplitude.
For a sine, the RMS is 1/sqrt2 times the peak amplitude.
For a square wave, the RMS is equal to the peak amplitude.
For a sine, the RMS is 1/sqrt2 times the peak amplitude.
For a square wave, the RMS is equal to the peak amplitude.
It is always good advice to ask R. first.
For practical use of a signal generator I think it is advanteageous to calibrate
all waveforms for the same peak level output. This comes in handy if you switch
around when investigating an amplifier. For "daily life" the oscilloscope picture is
important, RMS is not.
For practical use of a signal generator I think it is advanteageous to calibrate
all waveforms for the same peak level output. This comes in handy if you switch
around when investigating an amplifier. For "daily life" the oscilloscope picture is
important, RMS is not.
Thanks for the advice. I understand what you say. And it’s really good to learn new things. I had no idea about the square pp - rms relationship.
But is there any suggestion regarding the low scale issue? As you can see from the original post (incl. service manual) my problem is rather practical. Not crucial, but still a problem.
But is there any suggestion regarding the low scale issue? As you can see from the original post (incl. service manual) my problem is rather practical. Not crucial, but still a problem.
Without going into detail what you see in the lower scales may be an artifact in your set up.
The generaor output is usually switched as a resistor divider network. The resistors are fixed
most of the time, no adjustments. The network will not fail in general use and is not prone to
overload if you avoid exposing it to dc.
Apart from that your description is hard to understand and may be affected by translation and
typos.
The generaor output is usually switched as a resistor divider network. The resistors are fixed
most of the time, no adjustments. The network will not fail in general use and is not prone to
overload if you avoid exposing it to dc.
Apart from that your description is hard to understand and may be affected by translation and
typos.
Thank you As_Audio for the effort and input. I can live with my low scale problem. I will try some other cables and dapters and if nothing better hppen, that’s it.Tried to edit and improve my initial post as well.
Is english your native language, if we know where you live some forum "neighbours"
can surely offer assistance ..
can surely offer assistance ..
Athens, Greece. English is a foreign language for me- but have lived and worked in london, uk in the past.
I guess that they used to speak english in London (but today ?)
Had a quick look at the generator schematic, simple and straight forward british style.
Do not expect wonders of accuracy here.
It has a power amp output so that you can check speakers ..
Had a quick look at the generator schematic, simple and straight forward british style.
Do not expect wonders of accuracy here.
It has a power amp output so that you can check speakers ..
So many accents and languages there. Haven’t been there since 2010. Very simple and cheap old generator. I will also check with a computer program called REW. It has a very sophisticated tone generator. To tell you the truth i prefer simple test equipment, but i may try it just fir comparison.
calibration went well. Oscilloscope was needed just for the square.
The problem is the low sine scale. And this can’t be fixed with callibration. It’s more to do with cables, power supply and general age of the generator parts.
The problem is the low sine scale. And this can’t be fixed with callibration. It’s more to do with cables, power supply and general age of the generator parts.