|
Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders. |
|
Thread Tools |
19th Jan 2019, 4:06 pm | #1 |
Heptode
Join Date: Sep 2018
Location: Twickenham, London, UK.
Posts: 536
|
Nombrex 27 with distorted output
I recently purchased at reasonable cost a Nombrex model 27 signal generator. The seller said that it worked but that the output was "dirty". Looking on a scope the negative going half of the waveform is reasonable but the positive half has distinct distortion which is worse at lower frequencies. I have attached traces at 500KHz, 1.5MHz and 20MHz. My question is whether this is normal for what was a simple low cost instrument or is there a fault? If it is not normal is this a known fault? Needless to say that are lots of harmonics present in the waveform. I have also attached the circuit diagram, in my example the RF transistor is a 2G417 and the AF transistor is a 2G271B. I have also attached a photo of the internals. I have no experience of the tubular PC mounting capacitors used, are they reliable? Any thoughts welcome.
Alan
__________________
Alan G6PUB, BVWS |
19th Jan 2019, 7:46 pm | #2 |
Heptode
Join Date: Jun 2010
Location: Zurich, Switzerland
Posts: 538
|
Re: Nombrex 27 with distorted output
Alan,
I dont have that generator, but it`s a 40-45 years old product: I would search for cold solderingx, some GND contact problems, clean properly the switches... BTW: tubular Cs are reliable, Rf type capacitor, often with special TempCo-needed ie for tuned circuits_I wouldn`t touch these. Can you post full schematics/parts list please, than yours seems to differ from that of Radiomusem... rgds, Karl I assume, the higher frequncy signals are worser-right? Last edited by karesz*; 19th Jan 2019 at 8:09 pm. |
19th Jan 2019, 8:37 pm | #3 |
Heptode
Join Date: Sep 2018
Location: Twickenham, London, UK.
Posts: 536
|
Re: Nombrex 27 with distorted output
Thanks for your response Karl, I have already visually checked joints etc and checked switch contacts. The distortion is worst at low frequencies so going from left to right the oscillograms are 500KHz, 1.5MHz and 20MHz. The schematic and parts list above is all I have and my unit appears to conform to the schematic.
If anyone has one of these signal generators I would appreciate if they could see what the waveform looks like. Other than the distortion the generator works as it should do and for most tasks could probably used as is. Alan
__________________
Alan G6PUB, BVWS |
19th Jan 2019, 8:54 pm | #4 |
Heptode
Join Date: Sep 2018
Location: Twickenham, London, UK.
Posts: 536
|
Re: Nombrex 27 with distorted output
Karl, I have just looked at the entry on the Radiomuseum website and it appears to show two different versions of the instrument. I assume that at some point the circuit and layout were revised, I would guess that my one is the later version but I can't be sure.
Alan Edit: A note at the end of the Radiomuseum entry confirms first schematic 1962 and second schematic 1966
__________________
Alan G6PUB, BVWS Last edited by Vintage_RC; 19th Jan 2019 at 9:06 pm. |
19th Jan 2019, 9:52 pm | #5 |
Heptode
Join Date: Jun 2010
Location: Zurich, Switzerland
Posts: 538
|
Re: Nombrex 27 with distorted output
Alan,
I think taht your exemplar with tubular caps is more old -if they are original, what would assume- as these attached pic is. Anyway, ther are versions in the worled, you can check a video-Ok over model 41 too, but maybe help some for you...BTW; are thet transistors original ones? Check these site from finland_ he suggest some Transistor types for replacements. Hese S/N is 2167. I found that Phil had to play with an model31-so 5 years ago: he had an tubular C to change... Last edited by karesz*; 19th Jan 2019 at 9:58 pm. |
19th Jan 2019, 9:55 pm | #6 |
Heptode
Join Date: Mar 2009
Location: Portsmouth, Hampshire, UK.
Posts: 674
|
Re: Nombrex 27 with distorted output
I think those waveforms are quite normal, the output is rich in harmonics (= distorted).
I suspect that the apparent lack of distortion at 20MHz may be due to bandwith limiting by the 'scope. Jim |
20th Jan 2019, 1:14 am | #7 |
Heptode
Join Date: Sep 2018
Location: Twickenham, London, UK.
Posts: 536
|
Re: Nombrex 27 with distorted output
Thanks for the links Karl. I just checked the bandwidth of the scope and it is 25MHz so I think you are correct Jim.
Alan
__________________
Alan G6PUB, BVWS |
20th Jan 2019, 7:03 am | #8 |
Dekatron
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,637
|
Re: Nombrex 27 with distorted output
Check R value's but it's only a simple sig gen, it's never gonna be as clean as a "proper" sig gen.
Andy.
__________________
Curiosity hasn't killed this cat...so far. |
20th Jan 2019, 10:55 am | #9 |
Heptode
Join Date: Feb 2010
Location: Heysham, Lancashire, UK.
Posts: 665
|
Re: Nombrex 27 with distorted output
In my youth, you had “Standard Signal Generators” (SSG) such as those from Marconi, HP and the like, and “Modulated Test Oscillators” (MTO) such as those from Nombrex, Heathkit, or Trio.
SSGs had high levels of screening, controlled output levels using either output level meters or AGC, and calibrated attenuators. MTOs were more basic oscillators in a simple metal box with a pot to adjust (I hesitate to use the word “control” with the level of screening used) the output level. The Nombrex is battery powered, so if you need a really low level of signal for alignment purposes, you can move it to another room. Stuart |
20th Jan 2019, 8:23 pm | #10 | |
Rest in Peace
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
|
Re: Nombrex 27 with distorted output
Quote:
Al. Last edited by Skywave; 20th Jan 2019 at 8:25 pm. Reason: Add quote |
|
21st Jan 2019, 1:22 pm | #11 |
Heptode
Join Date: Sep 2018
Location: Twickenham, London, UK.
Posts: 536
|
Re: Nombrex 27 with distorted output
RF transistor is a 2G417. I have already done some checks on the biasing voltages around the RF transistor with a DVM. Results were a little strange, I suspect RF was influencing the DVM readings. I will do a few checks with my GEC Selectest. I don't want to start pulling out lots of components to fix what might be an inherent design limitation, I was hoping someone might have one that they could check the waveform on but no luck so far.
__________________
Alan G6PUB, BVWS |
26th Jan 2019, 4:12 pm | #12 |
Heptode
Join Date: Sep 2018
Location: Twickenham, London, UK.
Posts: 536
|
Re: Nombrex 27 with distorted output
I have carried out a few more measurements on the Nombrex 27 looking at the biassing of TR1. I used a scope with DC coupling set and measured relative to the positive rail. TR1 base was at 0.5V and TR1 emitter was at 0.2V, both with superimposed RF. A few simple sums base on the component values suggest the base should be at 1.4V and hence the emitter should be at around 1.1V. I wondered if the fact that the circuit was oscillating was causing the apparently odd results as I have observed this before. I stopped the oscillation with a 0.1uF ceramic cap from TR1 collector to ground and then found that all bias voltages were as expected. Regarding the nature of the distortion I found that the waveform continually altered as you tuned across the ranges. My conclusion is that the instrument is almost certainly working normally and is what you might expect from a simple circuit that covers a wide frequency range. Looking at the component values it seemed to me that the power rail decoupling capacitor is rather low at 0.003uF for an instrument that operates down to 150KHz (reactance 350 ohms). I might try increasing it but can't go too far as ultimately it will affect the audio modulation.
__________________
Alan G6PUB, BVWS |