Recommended sweep generator/oscillator/wobbulator.

[djsbriscoe]

Can anyone recommend a SMALL (I don't have much bench space) RF sweep generator/oscillator/wobbulator? I bought a use Marconi 2022E Signal generator a while ago which whilst useful it doesn't have a sweep facility. I have a Raspberry Pi wobbulator kit waiting to be built but I'd like to get something that is NOT tethered to a PC/laptop. Thanks.

David

[chriswood1900]

Can you give some idea of what you want to use it on, or what frequency ranges you will need and an indication of budget. There are some great Wobbulators from leading suppliers but they might be a tad expensive.
Chris

[MrBungle]

Not sure there’s such a thing as a small one.

Some of the new arbitrary waveform generators are good for this. My main signal source is a Rigol DG1022Z. That’s will sweep 0-25MHz with a log x scale and has marker generator output. For anything above that I use a Marconi TF2015 which is good from 10-520MHz. The sweep input is driven from the DG1022Z.

AFAIK there’s no such thing as a cheap one though!

[Radio Wrangler]

Sweepers usually mean having the ability to sweep wide ranges.

Wobbulators usually are designed for narrow sweeps around a centre frequency such as for aligning IF filters in receivers.

Some instruments cover both abilities, but you need to be sure of which you need otherwise you risk getting the wrong one or setting out on a search for something less common than you needed.

You also need to consider what range of levels you need.

If you're stuck for space, you may want mod facilities so it doubles as a signal generator

David

[kevinaston1]

The smallest that comes to mind is the Philips PM5326, covering 0.1 - 125MHz.


Kevin

[Skywave]

Originally Posted by djsbriscoe

I bought a used Marconi 2022E Signal generator a while ago which whilst useful, it doesn't have a sweep facility.

In the past, I have used a 2022E to align the 100 kHz IF strip in an Eddystone EA12, by sweeping that I.F. amp. It's not easy trying to work out just how to get the 2022 to do that, using the Ext. Mod. facility, but it can be done. However, the handbook does provide some help. Just bear in mind that with an external mod. source (and with the 2022 appropriately set up), the deviation of the O/P signal will be a function of the amplitude of that mod. signal. (Standard text book FM theory). Finally, a 'scope at the end of the I.F. strip in needed, or look at the rectified r.f. after the detector.

Al.

[Craig Sawyers]

I have two, and both of them are capable of causing a hernia

[MrBungle]

TF2016 is an option as well if it's for broadcast band. That doesn't weigh as much but they're difficult to find by the looks.

I notice all the synthesized digital ones are a pain to sweep which is why I went semi-ancient analogue only and brand new digital

[Argus25]

Originally Posted by kevinaston1

The smallest that comes to mind is the Philips PM5326, covering 0.1 - 125MHz.

I would second this suggestion. The PM5326 is a fairly compact generator, really very well made and the beauty of it; is its internal frequency counter (Which based on an Intersil LSI IC). The generator is very well laid out inside, easy to service and was a leap forward from Philip's previous generation generators. They are excellent for aligning AM & FM radios. At this stage in their life the power supply elctros need replacing and they sometimes have troublesome in-line bridge rectifiers in the power supply section that need replacing, but other than those minor issues they are super reliable in my experience.

[G0HZU_JMR]

The other (small and cheap) option for a simple sweeper would be a cheapo MCU + DDS chip. You could get the whole thing in a fag packet sized box. Program start and stop and sweep speed into the MCU EEPROM and it would run standalone without a PC.

If you only need frequencies up to a few MHz then something cheap like the AD9833 DDS would be OK. Maybe use an AD8307 log amp as the RSSI detector that feeds the scope display.

Both the DDS and the MCU board (AVR? Arduino?) and the AD8307 RSSI board can be purchased ready made via ebay. Then just program it up to do a basic sweep and to dish out a start trigger and an end of sweep pip for channel 2 on the scope. This would help you calibrate the frequency display on the scope. The AD8307 logamp response is 25mV/dB over >>70dB.
It wouldn't be difficult to add an attenuator to set the level output as required.

[Argus25]

Originally Posted by G0HZU_JMR

The other (small and cheap) option for a simple sweeper would be a cheapo MCU + DDS chip.

The only issue I see with this suggestion of building test apparatus, the most difficult thing is missing from the project, all the appropriate RF screening required in the physical construction of a unit, and the RF attenuator. In other words the electro-mechanical design of the project, that takes it beyond some pcb's on the bench.

It important for any RF generating device for lab or test purposes, that it has has a properly made and shielded calibrated attenuator to reduce the known signal level in a controlled and known manner down to a level below which the radio under test can hear. It is not much of a test instrument without that. When you buy a pre-made generator like the PM5326, you get all that included and a great accurate frequency counter as well, that is why its hard to beat.

I made a generator with exactly the same output level as the PM5326 (copied it), and with a switched rather than linear attenuator. Even then there was a lot of metalwork/construction and the frequency coverage is only 200kHz to 2MHz. Just for AM radio alignment. The PM5326 goes to 125MHz and does sweeps to align FM radios. You can see my unit for AM radios here:

http://worldphaco.com/uploads/The__H...nsanalyser.pdf

So what I am suggesting is that whatever the active device/s were to generate and wobbulate the RF, it requires a metal housing, careful planning and shielding and the attenuator (even if a digital one) and its a lot of work to build a lab test generator that is any good. So the small or cheap notion is really out the window I think. Its like the problem of making a good RF probe too.

[G0HZU_JMR]

The problem is that it isn't clear what the intended use of the sweeper is or what the frequency or levelling requirements are?

I have a Raspberry Pi wobbulator kit waiting to be built but I'd like to get something that is NOT tethered to a PC/laptop.

Based on this I suggested an option to 'untether' the DDS away from the Raspberry Pi and make it standalone. This could be done with a tiny MCU chip and a DDS that would cost a couple of quid.

The Raspberry Pi wobbulator looks like it consists of an unscreened Chinese DDS board on an unscreened carrier PCB sat directly on top of the unscreened Raspberry Pi. My untethered alternative is no worse than this?

I mentioned the AD9833 DDS as this would be cheap and adequate for use across LW and MW. At these frequencies, it isn't difficult to get good isolation. Just put the DDS in a metal box if you ultimately want to attenuate the levels down to a few microvolts.

The real issue with my suggested option is that it requires some construction work and also few people will want the hassle of writing their own code for the MCU chip. Otherwise, all the bits can be bought from Farnell and it could be up and running in a few days for a low cost.

Note that quite a few (lab) RF sweepers don't have an internal attenuator but sometimes this is offered as an option. Many of them run with a very restricted level range somewhere around 1mW.

[G0HZU_JMR]

I'd also advise caution when looking at the specs of some lab sig gens that offer 'sweep modes' when reading through the glossy datasheet.

I have several classic lab (LF through UHF) RF sig gens here that boast the ability to sweep but often the sweep speed is painfully slow once set up as required. I haven't spent much time exploring the limits of the sweep modes in my generators but this is because the sweep mode is so clunky in all of them. It's also worth checking if there are any limitations to do with the minimum or maximum sweep range they can cover at a chosen centre frequency.

[MrBungle]

TF2015 is like that. It oscillates if you sweep it too fast.

[G0HZU_JMR]

It's a long time ago now but I remember experimenting with my Marconi TF2015 sig gen and my old Tek 585 scope back in the 1980s. The 585 had a (huge!) ramp output voltage which could be scaled/shifted down and used to sweep the TF2015 in line with the scope trace. However, it wasn't a great setup and for narrowband stuff I think I ended up feeding a low frequency (2 or 3 Hz?) sinewave into the 'ext mod' at the back of the 2015 (and also the horizontal sweep port on the scope) and I used 'ext FM' on the TF2015 and XY mode on the scope. I can't remember now, but I may have done this with my old Solartron A100 scope as well. It was possible to get a wobbulator effect on the scope like this but it wasn't that great and it was very fiddly to set up. The TF2015 doesn't have much FM deviation range even in x10 mode but I recall it was just enough.

[Argus25]

Originally Posted by G0HZU_JMR

Note that quite a few (lab) RF sweepers don't have an internal attenuator but sometimes this is offered as an option. Many of them run with a very restricted level range somewhere around 1mW.

Yes, that is why I like the PM5326, it has a linear attenuator. It is fully shielded in an enclosed housing, it is a very special part. I think Philips probably made it. When I saw some of these attenuators NOS on ebay a few years back I got some for projects. They have an external silver plated body and a 6mm control shaft. They maintain a uniform 75R output impedance and input impedance, so you can drop them into a 75R line without an impedance bump, they seem to act like the linear equivalent of a ladder attenuator, but I have not checked the frequency performance above 125 MHz. I have never pulled one apart to figure out how they did it. Most likely they are fine at UHF because a similar one is used in the PM5519 pattern generator as I recall.

You can see the body of this sort of attenuator in a PM5519 photo:

https://www.google.com.au/search?q=p...Xc0vY_ld8044M:

[m0cemdave]

Some discussion of those attenuators in this thread:
https://www.vintage-radio.net/forum/...d.php?t=152826

[Argus25]

Interesting, it looks like they are rated to 1.25 GHz !

I saw this spec:
INSERTION LOSS @ 10 MHz = 3.4 dB, RISING TO 5 dB @ 350 MHz, AND MAXIMUM IS 8 dB @ 1.25 GHz. MAXIMUM ATTENUATION REMAINS CONSTANT OVER THE RANGE FROM 10 MHz TO 1250 MHz.

[PJL]

My sweeping history goes:
1. Made a simple wobbulator for a CR100 by hacking an AM radio oscillator and adding varicap diodes. The sweep was provided by the scope and this was perfectly OK for the job.
2. Bought a 'project' Polyskop for £5 and spent 15 times that on manuals and parts but this has a 75 ohm input impedance so not ideal for domestic/amateur sets. It works but is so big and heavy it stays in the garage.
3. Picked up a 'project' TF2370 spectrum analyser and repaired it. I bought a low loss FET probe for it but would still need an attenuator if I wanted to sweep a VHF radio. It is also big and heavy but more useful than the Polyskop.
4. Bought a TF2022. Used it to increase the bandwidth on a mono FM tuner so it would work with an external decoder. Set FM modulation to +/-100kHz and external modulation. Connnected scope sweep output to TF2022 modulation input and scope probe to demodulated output. Worked a treat.

Unless you can afford near new kit or like a serious challenge repairing 'project' test equipment, my recommendation would be to avoid overly complex test equipment as 95% of your time will be spent keeping it going.

PS: I have a Tektronix 454A scope (old, but exceptionally good) and not all scopes have the required features.

[G0HZU_JMR]

Yes, I can remember having a lot of fun trying various things with the TF2015 to lash up a wobbulator. I didn't get great results though. For example, I didn't get very good results in terms of linearity of the sweep because I used a crude AF oscillator to drive the scope horizontal input. But I found it didn't really matter for the stuff I was doing. However, for radios that have adjustable filter coupling I guess you will need an accurate frequency scale on the scope.

The PM5326 sounds to be tailor made for stuff like this but I'm not sure how easy it will be to find one if it is needed asap. I've never seen one or used one but I guess they pop up from time to time on ebay etc.