HP3314A - problem with external modulation.

[WaveyDipole]

I have been working on an Advance type 63A signal generator and thought to compare its FM output to the modulated output from the HP3314A. I haven't used these modulation inputs on the HP3314A before but a glance at the manual suggests that simply connecting the signal to the appropriate input should be all that is required in order to generate a modulated signal. Unfortunately neither of the modulation options seem to be working and I am wondering whether I need to do something additional to enable the inputs?

Here is what I have discovered:

- the reduce input LED stay on the AM input stays unlit
- the AM LED does not illuminate when FM/VCO are de-selected
- the signal applied to the AM inoput does not modulate the output
- the FM/VCO button illuminates the appropriate LED, but the signal applied to the FM/VCO input does not modulate the output in any way for either setting

Output signal at 1MHz with an amplitude of 100mV. Input at 1kHz with about 50mV of amplitude for 50% modulation.

To all intents and purposes the whole modulation section seems inoperative.

Does the modulation section need activating somehow, or is this likely to be a fault? There are two separate inputs and three modulation functions neither of which are working so any fault would have to affect all of these.

Any views would be appreciated.

[Humber888]

I have never played with a 3314A so the following may be totally invalid. However, getting the internal mod to work on my 3325B is a bit sneaky at first. On the back there is a 'Mod Source Out' BNC that has to be linked to one of the 'Mod In' BNCs via a short cable. Then you still seem to need one more button press than you thought to get the modulation onto the main output. Why they couldn't get a single button press to turn the appropriate AM/PM mod on beats me.

[mhennessy]

I've just tried mine, and it seems to be pretty straightforward.

The way the switching logic works is that the modulation button cycles between off, FM and VCO. If you want to enable AM, you have to press the blue "shift" button first. To turn off AM, you again press the blue button first. This logic enables you to combine AM and FM (or VCO) if desired. The power-on default is no modulation, so no LEDs.

(Sorry if that's stating the obvious, but from what you've said, that might part of the problem)

First I tried AM. Set the 3314A to 100kHz and applied about 150Hz sine wave from another generator. Pressed the blue button followed by the modulation button, and the AM and "Reduce input" LEDs both came on. Adjusting the level of the other generator, the threshold of this LED is about 700mV RMS. Looking at the output of the 3314A with a 'scope, the waveform was amplitude modulated as you'd expect.

Next FM - I turned off AM in the same way I'd turned it on, then moved the signal cable from the AM input to the FM input, and pressed the modulation button to light the FM LED. That's all I needed to do. The threshold for the "Reduce input" LED is the same as before - around 700mV.

You'll need more than 50mV going in to achieve 50% modulation. Remember, there's no relationship between the input modulation signal and the set amplitude of the 3314A generator output amplitude (sorry again if I've misinterpreted what you've said and stated the obvious).

I've had mine for quite a few years now, but if I'm honest, I've not used it as much as I thought I might. The waveforms are high quality - probably better than my preferred Wavetek 162, but when it comes to function generators, I like knobs! The fan noise is annoying, and it doesn't half chuck out some heat once it's been on a while. I thought I might use it for radio work (because the Wavetek lacks AM), but I've found the Philips PM5134 to be better. It has a fan, but it's inaudible, and it has an internal modulation oscillator and can do most of what the 3314A can do - with knobs

Hope this helps,

Mark

[WaveyDipole]

Mark, agreed regarding the heat and noise. Makes it rather difficult to peak anything audibly.

I proceeded as you suggested and got the required result. It was a combination of not being aware of having to use the blue button and insufficient signal amplitude. As you pointed out, pressing the blue button is necessary to enable the AM modulation mode and a signal well above 100mV. After pressing the blue button the AM light comes on. The Reduce Input light also comes on if I turn it up too high and I can see the typical AM modulated signal on the oscilloscope screen. As you mentioned, both AM and FM modes can be turned on at the same time.

It seems to take a significant level of signal (around 800mV RMS - 1.5vpp - or so) to get a visible deviation on the FM modulation mode. Rather less than that on VCO, but at least they both work. The signal looks the same as on the Advance 63A so it does appear to be producing the correct output. Illustations of an FM trace I have seen online seem a bit mis-leading e.g.:

https://upload.wikimedia.org/wikiped...mfm3-en-de.gif

It does give the general idea, but the signal does look somewhat different on the scope.

Anyway, I can now confirm that these functions do work on my HP3314A, I just needed to learn how to use them. Thanks for the steer.

[mhennessy]

That's good news

I'm glad you haven't got to dive in there as it's a pretty complicated beast. In case you haven't already looked, there's some internal pictures here: https://www.eevblog.com/forum/testge...n-explanation/

Have you tried the Easter Egg yet? https://www.youtube.com/watch?v=H90d6sPUF6A

[IanG3XYV]

WD might recall he acquired his 3314A from me some time ago. I never used it in AM or FM mode so didn't feel able to contribute anything here but I'm delighted Mark was able to show the way.

Ian

[WaveyDipole]

Originally Posted by mhennessy

That's good news

I'm glad you haven't got to dive in there as it's a pretty complicated beast. In case you haven't already looked, there's some internal pictures here: https://www.eevblog.com/forum/testge...n-explanation/

Have you tried the Easter Egg yet? https://www.youtube.com/watch?v=H90d6sPUF6A

Yes, found out about and tried the Easter egg some time ago and yes, it does work and it must play for a good three or four minutes. Must have taken someone quite some time to program that in including harmonies!

Thanks for the pictures. That does indeed look like a quite densely populated case with a number of separate boards in layers. Naturally the board for the Modulation circuit would be right at the very bottom.... So yes, I'm glad I didn't need to delve into there!

Originally Posted by IanG3XYV

WD might recall he acquired his 3314A from me some time ago. I never used it in AM or FM mode so didn't feel able to contribute anything here but I'm delighted Mark was able to show the way.

Ian

Ian, yes I remember picking it up from you. It has been a vary useful tool and has served for a number of projects. In fact I used the sweep mode just the other day to visually align the IF cans in the Kriesler radio that I have been recently working on. Thank you for your post.

I did check a couple of times before I posted and just checked again to be sure, but the Operating Manual does not seem to mention pressing the blue button. Neither does the Service Manual, although that is naturally more focussed on what needs doing rather than how to do it. Of course anyone paying close enough attention would probably have noticed that the letters AM were written in blue...

Hindsight is a wonderful thing...

[mhennessy]

Oh dear!

Earlier I noticed a faint but unmistakeable smell in the workshop...

Bench power removed, I set about tracing it, and the HP 3314A was the culprit. The X-cap in the Schaffner mains inlet filter was slowly letting go. Fortunately I got to it before there was too much mess - all limited to the rear panel. Luckily, that comes out easily, making the clean-up reasonably easy - that's all done now.

Looking at the pictures in the manual, it appears that earlier models used a standard IEC inlet which is the same size as an unfiltered inlet. Replacements are widely available and cheap, and the job looks straightforward - if you have this type, I'd definitely recommend adding this to your to-do list.

Mine is one of the newer versions, with the soft-touch rubber keyboard rather than the classic "clicky" switches. The mains inlet combines the voltage selection within it (the voltage selection switches are separate in earlier models), and therefore must be replaced with the same unit.

The good news: the exact module - incorporating the mains voltage selection - is available. But the bad news? RS charge £52 for them!

There are cheaper equivalents, and there are probably cheaper suppliers too; I'll research that later in the week, but in the meantime, I hope this serves as a useful reminder/warning to us all. The mains filter components are ahead of the front panel switch, so are powered all the time my bench is switched on. I've no idea why it decided to do this today but if nothing else, it's a spooky coincidence!

[WaveyDipole]

Ooh! Sorry to hear that.

However, thank you for the heads up. Mine is of the "clicky" button variety so no doubt older than your unit. I will definitely add an inspection and pro-active replacement to my todo list.

[IanG3XYV]

Glad you’re having some good use out of the 3314A. The mains inlet filter seems to be an occupational hazard of this vintage of test gear. I’ve had to change them in a piece of Philips gear and HP. Not difficult to replace fortunately.

[mhennessy]

Yes, they are a pain - I've lost count of the number I've changed. I've stopped rescuing them from scrap kit (unless it's relatively new stuff, which is actually depressingly common in broadcast these days). The last time I fitted a "rescued" filtered inlet, it lasted about 15 minutes before letting go! I confess that sometimes I just fit a standard unfiltered socket, which I know is naughty, but when you're in a hurry...

Anyway, and with apologies for "thread-jacking" (I'd understand if this was split off into a separate thread), I have just ordered the correct replacement for my late 3314A.

As mentioned, the inlet is a combined unit that contains a fuse and a voltage selection switch, and it snaps into the rear panel. It would be hard to fit anther type because space on the rear panel is limited - otherwise I would have considered a plate and a more conventional inlet. But my example is practically mint, so I couldn't possibly consider a bodge, even if the costs are a bit stiff!

The type fitted is a Schaffner FN370-2-22.

RS have them, but in single quantities they cost around £50: https://uk.rs-online.com/web/p/iec-filters/4659347/ - that really is too much. You can get them at a sane price if you're willing to buy a box of 40, but clearly that's no use to me.

There is a close match: https://uk.rs-online.com/web/p/products/8826784/ - this is an FN372-22-2, and costs around £20 in single quantities. That's more like it...

The only difference is this unit has 2 mains fuses rather than just 1. This would probably have been fine, as I could have simply ignored the extra fuse holder - and it's possible I might be able to pull out the bits of metal that connect to the extra fuse. They're on back-order though - 2 weeks at time of writing.

But Mouser have the correct type in stock for pretty much the same price. Delivery is an extra £12, but only if you spend less than £33. I was able to find a few other parts that I've been meaning to order to get the price up to that threshold, so hopefully I'll be able to put the 3314A back together before too much longer.

Oh - one difference between the original and the new replacement: the voltage selector "wheel" has a pattern of cams to make the appropriate switch contacts. The original has 100, 120, 220 and 240V positions. The modern versions have 100, 120, 230 and 240V positions.

I'm sure the cam arrangements are identical (but will check very carefully!) so I'm hoping I can simply use the old wheel with the new inlet. But if not, it's hardly the end of the world...

Anyway, I hope this information is of use to someone with a late HP 3314A, or indeed, anything else that uses this combined inlet with filter/fuse/voltage selector. When it arrives, I'll take some pictures of the repair.

Mark

[WaveyDipole]

I had a look inside my 3314A today. The filter present in this unit is not made by Schaffner but is stamped Corcom, Hong Kong. I couldn't see the model number but I haven't taken the rear panel off yet. If it originally had a Schaffner EMC filter then perhaps it has been replaced? However, the three calibration seals had not been disturbed so maybe this was done by a calibration house prior to Ian's ownership of it? The next calibration due date is apparently 2009, so if so, then it was some time ago.

I had a quick Google and it turns out that Corcom are a known EMC filter brand stocked by the likes of CPC Farnell and Mouser so it would seem the replacement (assuming it is one) is by a reputable brand. I see no obvious muck on the inside so it doesn't look like the old one had let go, but I haven't dismantled it far enough to get a really good look. Perhaps it was a pro-active replacement or just cleaned up very well.

Will this Corcom be OK for a while yet - or is it best replaced?

[mhennessy]

Hi again,

This thread has just reminded me that I promised to upload some photos of the recent repair.

Hopefully all self-explanatory, but I ought to mention the metal brackets that are added to the inlet assembly - these need to be transferred to the new part.

Note the uninsulated mains connections - be careful inside one of these

And yes, as discussed above, I transferred the voltage selection cam from the old to the new. They are identical apart from the printing. But because I was being extra-careful about that, I did test it by applying 100, 120, 220 and 240 mV from another function generator and checked how many millivolts I got at the secondary winding. Of course, a variac and lamp limiter would have been equally good, but that would have involved a bit of heavy lifting and clearing some more bench space, whereas the other function generator was already in place

That fan is a thing of beauty, by the way. Machined from solid aluminium, and wonderfully smooth and free bearings. Just a shame it's so noisy!

Finally, the last picture shows the earlier model, which has a more straightforward arrangement - in terms of ongoing maintenance, at least. Picture credit: https://www.sglabs.it/en/product.php...t-3314a&id=569

Mark

[Radio Wrangler]

The clicky key switches, with the hard plastic buttons are called Bill West switches after their designer. They were designed primarily to be cheap. They have two plastic parts, a very thin shell which houses the thing and a little strip of thin stainless steel which acts as the spring and creates the tactile snap. The moving part is white plastic and very minimalist, it's hollow to let an LED shine into the keycap for an illuminated dot if needed. There is a tiny gold plated metal sheet formed into a group of contcts, heat staked onto the moving part. This bridges gold plated pads on the PCB. The shell has lugs that get heat staked to the PCB.

Over time, the spring strips lose their action, gold wears through on the PCB and tarnish makes operation iffy.

Never throw away any PCBs with these switches on them unless all switches are dead. The heat stake acn be cut off with a scalpel and good switches saved to rescue other gear.

They've lasted longer than ever expected, but they were uncharacteristically cheap and nasty for HP

VCO mode means that you get a specified frequency/voltage characteristic. What the frequency does reflects what you do with the input voltage.

AM and FM modes allow you to set the percentage mod and or deviation. So long as you input a level between certain bounds, it will take care of things and give you the deviation or percentage you enter on the keyboard.

I've little experience of the 3314A (A loveland division product, by the way) But I did get the job of redoing the frac-N synthesiser in the 3325 when the frac-N ASIC wenobsolete and stocks were running low.

David