Wavetek Model 271 Function Generator Restoration & Calibration

[xrunner]

This is my next project - a Wavetek Model 271 Function Gen. I got it off Ebay for only $80. I've used one before but it's probably been 25 years. I haven't seen much on the EEVBlog regarding this model, so I thought it'd be interesting to some of you. Will be posting much more as I get into it. Nice pretty VFD display, still very bright.

[xrunner]

Checked out a few waveforms and it seems to be working. One issue is a low battery message, as so many of these older pieces of kit used batteries that had to be replaced to hold settings. This particular battery was totally dead - 0.0V and after an exhaustive search I found it is no longer available / no substitute. It's a BR-1/2A. I found batteries close in size but they would be too long to fit in the battery holder, so I plan on replacing it with a 3V battery with solder tabs - maybe a coin cell? Any ideas welcome.

A few more pics. Front panel is a membrane type keyboard, dirty but that should be simple to clean off. Many of the metal panels were date-stamped, which is neat. It's chock-full of electronics and boards to get the job done, but seems to be well-made and designed.

Lots of dust inside ...

[Andy Watson]

It's a BR-1/2A. I found batteries close in size but they would be too long to fit in the battery holder, so I plan on replacing it with a 3V battery with solder tabs - maybe a coin cell? Any ideas welcome.

I replaced mine with a Varta CR 1/2AA. It's slightly longer than the original but it can be persuaded to fit.

[xrunner]

It's a BR-1/2A. I found batteries close in size but they would be too long to fit in the battery holder, so I plan on replacing it with a 3V battery with solder tabs - maybe a coin cell? Any ideas welcome.

I replaced mine with a Varta CR 1/2AA. It's slightly longer than the original but it can be persuaded to fit.

I looked at that battery and did notice it was several mm longer. Not sure if I'll use that or a battery with leads already attached and just mount the battery on the side, or even just remove that battery holder since that battery is no longer made and there is no substitute.

[Nuno_pt]

xrunner,

In my opinion I would put a new holder in place that fit there for an easy obtainable battery.

[Andy Watson]

I looked at that battery and did notice it was several mm longer.

Yes, it's not the best of fits! I think the previous owner had already "adjusted" the holding bracket. It is quite solid once it's in position.

[xrunner]

In my opinion I would put a new holder in place that fit there for an easy obtainable battery.

That's an option to consider yep.

Yes, it's not the best of fits! I think the previous owner had already "adjusted" the holding bracket. It is quite solid once it's in position.

That was the first battery I was considering.

I might unsolder the batt. holder and try to make some new accurate bends with pliers so as to accommodate a longer battery. Might as well try since there's no battery going in it again anyway. 

If that doesn't work I'll put in a new battery holder.

[edpalmer42]

I have a Wavetek 278 that has the same battery issue.  The previous owner replaced the battery with a 1/2AA cell.

I see in the manual that the battery provides backup for "up to  6 months (typically 1 to 2 years)."  The 1/2A battery has a capacity of 650 mah while the 1/2AA has a capacity in the range of 850 - 1000 mah for a potential lifetime of 2 - 3 years.  I haven't done it yet, but I'll measure the current drain of my unit to see just how much it really draws.

If the current drain is similar to the typical values mentioned above, I'm not comfortable with the idea of soldering in a battery that has to be changed that often.  Buying a new lithium battery that often doesn't seem like the best plan either.  I might install a battery holder somewhere that holds 2 AA alkaline batteries.  Cheap, easily replaceable and, if it leaks, the damage will be seperated from the board.

Ed

[xrunner]

I have a Wavetek 278 that has the same battery issue.  The previous owner replaced the battery with a 1/2AA cell.

I see in the manual that the battery provides backup for "up to  6 months (typically 1 to 2 years)."  The 1/2A battery has a capacity of 650 mah while the 1/2AA has a capacity in the range of 850 - 1000 mah for a potential lifetime of 2 - 3 years.  I haven't done it yet, but I'll measure the current drain of my unit to see just how much it really draws.

If the current drain is similar to the typical values mentioned above, I'm not comfortable with the idea of soldering in a battery that has to be changed that often.  Buying a new lithium battery that often doesn't seem like the best plan either.  I might install a battery holder somewhere that holds 2 AA alkaline batteries.  Cheap, easily replaceable and, if it leaks, the damage will be seperated from the board.

Ed

Thank you very much for that info. My plan for the moment is to disassemble the entire unit (I would have anyway for cleaning) and unsolder the battery holder and try to re-form it with tools so it will fit the 1/2AA cell better.

I have taken out the logic boards on one side now. I noticed that the top board was designed to "flip-up"  for maintenance on two hinges. However, the two hinge pins were not in their respective holes  , so the hinges would not work. Hopefully you can see that in the pictures. Anyway, the boards came out just fine, but I noticed that even if I put the hinge pins in the holes so it would have flipped up (or tried to), that a shield was in place that prevents it from flipping up. The shield bumps against the frame way before the board can move up very much. If you take the screw out of the shield so the board can flip up - the hinge support falls off anyway so why is it there. Something of a mystery. Either it came from the factory made that way and they realized the design didn't work, or ... or I don't know what.

[edpalmer42]

I measured the battery drain on my unit.  Looks like it's not going to be a nuisance after all.

I measured 12 ua @ 3V0 and 13 ua @ 3V6.  Not impressive by today's standards, but within spec.  The battery is powering two HM6116LP-4 CMOS rams.  Maximum drain for them is 30 ua each @ 3V0 in standby mode.  Therefore, a 650 mah battery should last for over 6 years.  Wavetek's 'typical 1 - 2 years' is based on the maximum drain of 30 ua each.

So the dead Tadiran TL2150 cell in my unit (1/2AA rated for 1 ah) would have lasted for over 9 years!

By the way, it looks like the models 270, 271, and 278 are all variations on a theme.  Start with the 270 and add more features for the 271 and even more for the 278.  I didn't find a manual for the 271, but there are manuals for the 270 and 278 available online.

Ed

[xrunner]

I measured the battery drain on my unit.  Looks like it's not going to be a nuisance after all.

I measured 12 ua @ 3V0 and 13 ua @ 3V6.  Not impressive by today's standards, but within spec.  The battery is powering two HM6116LP-4 CMOS rams.  Maximum drain for them is 30 ua each @ 3V0 in standby mode.  Therefore, a 650 mah battery should last for over 6 years.  Wavetek's 'typical 1 - 2 years' is based on the maximum drain of 30 ua each.

So the dead Tadiran TL2150 cell in my unit (1/2AA rated for 1 ah) would have lasted for over 9 years!

Thanks Ed. Good info. I got busy with the battery holder, and managed to remove it, but not without a fight. It was not only held by two solder pins, but also by a sort of rivet joint. I had to carefully drill out that rivet. I removed the holder because I have several ideas for modding it, and even if that doesn't work out, I'll replace it anyway with a new size for a new battery. But I want to try modding it first.

By the way, it looks like the models 270, 271, and 278 are all variations on a theme.  Start with the 270 and add more features for the 271 and even more for the 278.  I didn't find a manual for the 271, but there are manuals for the 270 and 278 available online.

Ed

I did find a "Wavetek 278 Operators Manual, which covers the 271. Never found one specifically for the 271 though.

[TiN]

Holy trimmer batman. There are many of those in this box.
Thanks for teardown and sharing!

[edpalmer42]

Holy trimmer batman. There are many of those in this box.

Yeah, and there's something a bit weird about that.  Notice that the trimmers are square (Bourns?) 10 turn.  But the silkscreen outline is round.  My 278 has single-turn round trimmers where xrunner has 10-turn square ones.  I only see one round one near the yellow connector.

Xrunner, does it look like those trimmers are original or have they been replaced?  A couple of mine are broken.  The round ones are still available from Digikey.  Now I'm not sure what type to use.

Ed

[xrunner]

Holy trimmer batman. There are many of those in this box.
Thanks for teardown and sharing!

Yea a tweakers paradise. Going through the cal should prove interesting. 

Yeah, and there's something a bit weird about that.  Notice that the trimmers are square (Bourns?) 10 turn.  But the But the silkscreen outline is round.  My 278 has single-turn round trimmers where xrunner has 10-turn square ones.  I only see one round one near the yellow connector.

Xrunner, does it look like those trimmers are original or have they been replaced?  A couple of mine are broken.  The round ones are still available from Digikey.  Now I'm not sure what type to use.

Took a look at the board after you mentioned that, looked at the back carefully with a magnifying glass. In my opinion it hasn't been touched since it left the factory. Don't know why they decided to start using those 10 turn pots in those positions.

I took the battery holder to the garage and grabbed the vise grips. The CR1/2AA battery has a length of 0.99". I had a BNC termination that was about that length so I used that as a dummy battery. What I did was make some new bends in the end holders. You can see the new bends in the pic. I also added a dollop of solder on the neg. terminal to help make up the distance. I believe it will fit, and I have the battery on order.

[xrunner]

More disassembly.

Pics for reference in case anyone needs to see them someday. It's neat that there are date stamps on almost every piece of metal.

[xrunner]

Next up is a keyboard issue. All of the keys work, but the +/- key does not "click" like it's supposed to. The reason is it's obvious that key took a hit from some pointed object, as evidenced by an indentation in the front panel plastic. I'm going to be attempting a repair of the metal membrane under that key.

More tomorrow ... 

[xrunner]

So here we have the 271 front panel disassembled. I have notated the "ding" in red in the picture. An object hit the front panel, imparting a dent in the plastic and denting the metal membrane switch. The switch still works, but it doesn't have the nice tactile feel it should have. It's interesting that the ding is almost directly in the center of the switch, almost like someone tried to push the switch with a metal pointy thing.

I have an idea how to fix it, which I'll be working on today.

[edavid]

By the way, it looks like the models 270, 271, and 278 are all variations on a theme.  Start with the 270 and add more features for the 271 and even more for the 278.

They also made models 273 and 275.

I think the lineup was:

270 basic FG
271 FG + pulse
273 synth FG + sweep
275 FG + sweep + AWG
278 synth FG + pulse

[PA0PBZ]

I have some sort of love-hate relationship with Wavetek. I don't like the looks of it or the user interface, but I do like the build quality and the silkscreened functional blocks on the pcb's.
I do have a 178 (no idea how that relates to the 2XX series) and 2 98's in mint condition. To be honest they don't get much use since I have a Rigol DG4000 on the bench.
They just occupy too much space and lack the modern interfaces like USB or Ethernet.

[edpalmer42]

They just occupy too much space and lack the modern interfaces like USB or Ethernet.

Years ago I realized that the equipment I was going to be interested in only had GPIB so I made a point of picking up a few PCI GPIB cards and cables when I came across them.  I make good use of them for collecting data from counters and DMMs in particular.

Ed

[PA0PBZ]

Years ago I realized that the equipment I was going to be interested in only had GPIB so I made a point of picking up a few PCI GPIB cards and cables when I came across them.  I make good use of them for collecting data from counters and DMMs in particular.

True, and I did the same but now I've almost replaced everything with "modern" stuff. only the counter still has a GPIB port. My PCI card didn't even make it into my latest workbench PC but I do have a USB-GPIB cable that I can use. Some of the old equipment that I use even has ethernet already, like the E4406 VSA which must be over 10 years old.

[xrunner]

I have some sort of love-hate relationship with Wavetek. I don't like the looks of it or the user interface, but I do like the build quality and the silkscreened functional blocks on the pcb's.
I do have a 178 (no idea how that relates to the 2XX series) and 2 98's in mint condition. To be honest they don't get much use since I have a Rigol DG4000 on the bench.
They just occupy too much space and lack the modern interfaces like USB or Ethernet.

I understand. I have modern stuff too, like a Rigol DG1032Z. What am I doing with this thing? Because it's a hobby. I enjoy restoring the "vintage" test equipment to near-new condition, and playing with it. This Wavetek is fun to use, it's sorta like programming a calculator. 

By the way, the repair of the keyboard is going well, should have some pics by tomorrow.

[xrunner]

You can clearly see the dent in the pics. The entire back of the keypad has a plastic stuck-on sheet. What I did was to cut an opening along one side of the damaged metal part. I then used a tool (in this case a soldering helper tool) to insert under the metal and "scrubbed" the underside hard enough to re-form it back to where it should be. After I did this, it worked again as it should. I then sealed the cut with a piece of clear packaging tape. I also worked on the plastic membrane to get that dent out, by using the back of my X-Acto knife. Now the switch works as good as new.

[edpalmer42]

The adhesive on most packing tape dries out after a few years.  3M Magic Tape (the stuff with the green tartan label) will last for a couple of decades or more.  I don't know if it matters for this application.

Ed

[xrunner]

The adhesive on most packing tape dries out after a few years.  3M Magic Tape (the stuff with the green tartan label) will last for a couple of decades or more.  I don't know if it matters for this application.

OK, I'll keep that in mind. Not sure if it's a big deal here as it's basically pressed tight back there anyway by the mechanical assembly.

[xrunner]

Found something interesting regarding the battery. Without the battery, if you set the thing to cycle through "stats" by pressing the Stat button, after a few minutes the machine will Beep and reset to start up settings. I thought something may have been wrong with it, but the manual states that it will check the battery status every 6 minutes and sets a flag. Apparently this flag will cause a reset when the stats are cycling. Before I got the battery I hacked in a power supply set to 3V and this stopped.

But, I did get the battery yesterday and it fits well in the battery holder I modded. Pay attention to the battery marking - even though it's clearly marked with a "+" on an end. I put it in the first time thinking it was made like other batteries, like an "AA" with the nipple "+" and of course I still got the error message. It's not - it's just the opposite. 

[Radio Tech]

Nice tear down. I have always liked Wavetek stuff. When I worked at a General Electric land mobile radio shop I used a 3000B for a while. That is when I fell in love with Wavetek.





EDIT for spelling

[Andy Watson]

Found something interesting regarding the battery. Without the battery, if you set the thing to cycle through "stats" by pressing the Stat button, after a few minutes the machine will Beep and reset to start up settings. I thought something may have been wrong with it,

I think that is actually  a bug in the hardware design. When I purchased my 271 it was described as faulty - resets every few minutes. The battery monitoring shares an IC with the reset circuit. The processor tests the battery every few minutes by imposing a load and checking the result. Without the battery there is sufficient crosstalk from the test pulse to trip the reset circuit. If you want to run without the battery you have to disable the comparator, U27A, to prevent the resets.

[xrunner]

I think that is actually  a bug in the hardware design. When I purchased my 271 it was described as faulty - resets every few minutes ...

Wonder how many people it gave fits to?

I want to remove the grungy old foam around the fan. That old foam gets really mushy. It's in the garage now with the third coat of Goo Gone on the foam. It's really a bear to remove this one.

Getting close to calibration, and the tweaking of many pots. 

As stated earlier, I have the manual for the Model 278. There are 3 boards in the 271 and 278 (the other models may have that many also). Two of the boards are the same - Microprocessor/PSU and Function Generator. The 278's 3rd board is called a "Synthesizer". The 271's 3rd board has to to with generating bursts, widths, and delays, and has more trim pots than the 278's third board board. Maybe the Model 278 3rd board "Synthesizer" does the same thing but differently, I don't know. The 278 "Synthesizer" board has only two adjustments, so maybe it does it more in software.

I was worried I couldn't do the procedure on the 3rd board since the manual doesn't show my board in the cal procedure list, but I realized that the internal cal process (which you can recall starting from memory 1001) is very nice and might state what pot I need to adjust if I went past the last common cal routine for both units (RCL 1028). Sure enough, it does keep going, and by stepping up one memory at a time, the display tells me which pot to adjust for the 3rd board, and what the delay and PW should be. Cool!

Another interesting thing, is that on that 3rd board, the pots have a green line of tamper proof material, which as far as I can tell, has never been cracked or touched. I guess that means it has never been cal'ed, but then, why don't the pots of the Function Generator board have any tamper proof agent on them? Any ideas?

 

[xrunner]

Got the fan foam cleaned up and re-done. That original foam was really hard to get off.

[xrunner]

Well, I was setting up for cal, and when I tried to remove a BNC from the Sync Out, the whole connector started to turn. So, now I have to remove the front panel so I can get a wrench on the nut back there. Gah! 

[Nuno_pt]

It's things like that when happens that make you see how good you're on disassembling and assembling the thing. :-D
 

[xrunner]

It's things like that when happens that make you see how good you're on disassembling and assembling the thing. :-D

Had to take the back off, because the power switch prevents you from sliding the front panel out of the way up or down, and the switch has a metal bar that goes aaaaaaaall the way to the back. But in order to get the  back off I had to remove the bolts for the GPIB, and to do that I had to use long nose pliers which sucks. There are two small nuts on that GPIB conn. and I just don't have any sort of wrench that I can use on them, nor will a small nutdriver work because of how close the nuts are to adjacent parts. I don't know how Wavetek assembled it, but they might have had a set of very thin wrenches. I need to look on Ebay for some sort of thin wrench set.

[Nuno_pt]

I hate when that happen, you get stuck in fixing something because you don't have the right tool to do the job,
Then you've to wait this the tools/pats get there to finish it.

[xrunner]

Got in some replacement rubber feet for the rear today, as the originals were hard as a rock and cracked. Finally set up for calibration. Printed out the cal steps so I can mark them off.

So far, the +/- 15V supplies were checked and found to be pretty far off. Supposed to be 15 +/- 0.002, but were off by up to 0.4 V. This has already corrected waveform amplitude inaccuracies.

Just for curiosity, I skipped ahead to check some frequency cal points (but didn't adjust yet). Here's how those sit right now -

Cal Value       Current Value

999 Hz          1002 Hz
1.000 MHz      0.989 MHz
9.99 MHz      9.978 MHz
100 kHz       100.22 kHz
999 kHz       1001 kHz
12.0 MHz     12.025 MHz

[tautech]

Had to take the back off, because the power switch prevents you from sliding the front panel out of the way up or down, and the switch has a metal bar that goes aaaaaaaall the way to the back. But in order to get the  back off I had to remove the bolts for the GPIB, and to do that I had to use long nose pliers which sucks. There are two small nuts on that GPIB conn. and I just don't have any sort of wrench that I can use on them, nor will a small nutdriver work because of how close the nuts are to adjacent parts. I don't know how Wavetek assembled it, but they might have had a set of very thin wrenches. I need to look on Ebay for some sort of thin wrench set.

Very common predicament. 

Much equipment with I/O ports have these securing nuts that are normally 3/16" hex and best solution is a high quality hex Tee wrench with a thin walled socketed end. Brands like Snap On come to mind. 

Failing ones considerable investment in such a nice tool it's easy to get a 3/16" socket for your favorite 1/4" nut driver and give it a good grinding or turn it down a little in the lathe for when tackling these tricky little buggers.

[tautech]

So far, the +/- 15V supplies were checked and found to be pretty far off. Supposed to be 15 +/- 0.002, but were off by up to 0.4 V. This has already corrected waveform amplitude inaccuracies.

Just as you'd expect. 

Just for curiosity, I skipped ahead to check some frequency cal points (but didn't adjust yet). Here's how those sit right now -

Cal Value       Current Value

999 Hz          1002 Hz
1.000 MHz      0.989 MHz
9.99 MHz      9.978 MHz
100 kHz       100.22 kHz
999 kHz       1001 kHz
12.0 MHz     12.025 MHz

I'd be tempted to further *fine* adjust supply voltages to see if these values ^^^ fall into spec. 
If they do It'll save you a heap of work.

[xrunner]

I'd be tempted to further *fine* adjust supply voltages to see if these values ^^^ fall into spec. 
If they do It'll save you a heap of work.

What? And ruin all the fun of calibrating it! No way. 

Going through it, now on step 9 ...

[xrunner]

 

Basically done, but I want to post a shot of it together with my previous project ... 

[xrunner]

So the project is done, and it's working well. Here's a shot of it next to my previous project.

[GreyWoolfe]

 

[xrunner]

Thanks! And thanks to all who posted and contributed in the thread - I appreciate it! 

[xrunner]

Now I got another model coming - a 278.

Well ... it was only $59, what did you expect me to do LOL?

[xrunner]

The 278 just arrived. Looks to be in excellent condition for it's age. Seems to be working fine, and even the battery is OK. The bottom panel was still sealed with a fairly old cal sticker. What a deal for $59. I'll be restoring it the same way as the other one, basically giving it a complete clean-up and calibration. I might post a few pics specific to this model, but that's it.

Suh-weet! 

[xrunner]

Looks like they either didn't have 24 pin sockets in 1987 (that's the year this Model 278 was made) or they ran out of them at the assembly line. 

[PA0PBZ]

Looks like they either didn't have 24 pin sockets in 1987 (that's the year this Model 278 was made) or they ran out of them at the assembly line. 

Or it was just cheaper not to buy the odd 24 pin sockets.

[edpalmer42]

I just checked my 278.  Date codes are late 1983.  My sockets are also in two pieces so I guess that was their standard.

[xrunner]

Or it was just cheaper not to buy the odd 24 pin sockets.

Yea could be, doesn't seem like such a big cost to buy the right socket, but who knows?

I just checked my 278.  Date codes are late 1983.  My sockets are also in two pieces so I guess that was their standard.

Thanks. The 271 had them soldered in I believe. I always remove and re-seat all socketed chips in these older units, so that's why I noticed it.

Oh and another thing - the 278 (which was manufactured in 1987) has no fan inside, but the 271 does ...  hmmm

[Andy Watson]

Oh and another thing - the 278 (which was manufactured in 1987) has no fan inside, but the 271 does ...  hmmm

I noticed that. My 271 does not have a fan. Curious!

[xrunner]

I noticed that. My 271 does not have a fan. Curious!

Hey Andy - I want to post a picture tomorrow and have you check something inside your 271 - 'cause inquiring minds want to know. Check back tomorrow ... 

[edpalmer42]

My 1983 vintage 278 doesn't have a fan, yet the manual, also dated 1983 says it does.     Looking at my unit, the bottom of the case has a rectangular grill rather than the square grill shown in the manual and it doesn't have any mounting holes for the fan bracket.  So mine couldn't have a fan.    Beats the hell out of me!

Ed

[xrunner]

Here's the area of the Model 271 where the fan connects - J9.

[PA0PBZ]

Geez, all that Wavetek talk makes me want to put mine on the bench and play with it

[xrunner]

Geez, all that Wavetek talk makes me want to put mine on the bench and play with it

Yea - it's probably a little lonely. Turn it on and look at the pretty blue-green display and generate a signal ...  .

[edpalmer42]

Here's the area of the Model 271 where the fan connects - J9.

Yes, and my 278 doesn't have a connector or silkscreen or unpopulated holes in the board.  It was never intended to have a fan even though the service manual says it does.

[PA0PBZ]

Geez, all that Wavetek talk makes me want to put mine on the bench and play with it

Yea - it's probably a little lonely. Turn it on and look at the pretty blue-green display and generate a signal ...  .

Well, I did. Looks like I forgot how big this thing is 

[xrunner]

Well, I did. Looks like I forgot how big this thing is 

Oh my ... that's a big one. 

Looking at the specs it states it can generate "haverwaves", something to do with the waveform starting at +90 deg or -90 deg. Never heard of or used that term in my entire life. Learn something new around here every day.