Author Topic: Function Gen with Tubes - Exact 240 - Now with Manual and Schem  (Read 11659 times)

0 Members and 1 Guest are viewing this topic.

Offline AF6LJ

  • Supporter
  • ****
  • Posts: 2901
  • Country: us
Re: Function Gen with Tubes - Exact 240
« Reply #25 on: October 07, 2015, 01:59:54 pm »
Looks to be built in or after 1966 the two transistors in the first batch of pictures have clearly visible date codes.

I love it, looks great It would be cool to see the schematic.
I am betting the power supply has a series pass tube regulator for the high voltage.

So they do!
I must admit I didn't go & look at the pix offsite,so couldn't blow it up big enough.
When I did I could see them too.

They could be replacements----OK,so I'm reaching here! ;D

I was going by the general look of the thing,plus the fact they used Germanium transistors.
As a specialist device they wouldn't sell that many,so a 1966 one might not look any different to a 1959 one.

Back in 1965,my then work had a self -tuning ISB transmitter which used PNPs as part of the motor drive circuit for the  autotune,(it was  already a few years old).

It was s/n (1),& we basically did the "debugging".
It kept killing the PNPs.

Ultimately,a Supervising Tech & a couple of Senior Techs redesigned the circuit to use Silicon NPNs---no more problems.
The EE said "well done" & went on his way!
Very cool...
I am guessing the two TO3 transistors are for regulating the filament voltage for all but the pass tube used in the high voltage regulator. That doesn't need to be regulated and that pass tube will have its own filament winding since the cathode of the pass tube will be at a high potential. As a general rule the filament to cathode breakdown voltage is on the order of 200V.

I love this old stuff, it's fun to see how they had to do things back in the day.. :)
Sue AF6LJ
Test Equipment Addict, And Proud Of It.
 

Offline TimFox

  • Frequent Contributor
  • **
  • Posts: 924
  • Country: us
  • Retired, now restoring antique test equipment
Re: Function Gen with Tubes - Exact 240
« Reply #26 on: October 07, 2015, 02:29:42 pm »
Kepco used to refer to tubes as "NPNs with built-in pilot lights".  They kept building highish-voltage stabilized supplies with 8068 (beam-power tubes, essentially high-voltage 6L6s)  pass devices until they were discontinued, since they could take far more abuse than an equivalent silicon device in laboratory applications.
 

Offline Cubdriver

  • Supporter
  • ****
  • Posts: 2295
  • Country: us
  • Nixie addict
    • Photos of electronic gear
Re: Function Gen with Tubes - Exact 240
« Reply #27 on: October 07, 2015, 05:29:50 pm »
Those are likely multi-section electrolytics in the cardboard cans.  I see some are also mounted on phenolic wafers, too, and from a look at the backside pic it appears that they may be isolated from ground.  In case you don't know (and not being a tube guy you may not), the outer can on the multi-section electrolytics is typically connected to the negative side of the cap.  In cases where the negative side cannot connect to ground, they're mounted using insulating spacers, and often covered with cardboard tubes (as you have here), or shrink tubing and an insulating cap on the top (as they are on some old HP stuff I have worked on).

You'll note that there are symbols near the terminals (a square, circle, triangle, 'D' shape, etc.), and on the side of the cap (may or may not be on the cardboard; if not it should be embossed into the aluminum can under the cardboard) there will be a list of the symbols, along with the associated cap section value - for instance, D 40UF 250V, (square shape) 30UF 350V, (triangle shape) 10UF 400V would be a 3 section cap, common negative, with a 40uF, 250V cap; a 30uF, 350V cap and a 10uF, 400V cap inside, connected to the terminals as indicated.  They can have multiple values, and differing voltage ratings all within the same body.

It takes some effort, but they can be restuffed with a bit of patience and a few tools.  Modern electrolytics are often MUCH smaller physically for a similar value, and one or more of them as needed can be put in the can after the original guts have been removed.  As it appears you've discovered from your online searching, there are several methods used to restuff them - some cut the can and epoxy it back together (which may be a good option for you as yours have the cardboard covers which should hide the cut); others undo the crimp at the bottom to remove the guts, then re form it with a tack hammer or the like,  I prefer the latter method, though it's time consuming.

I've done a few single section ones for an old HP DVM; here are some of the pics of the process:

Initial opening of crimp with knife blade:


Continuing to open the crimp with a small screwdriver (a small electronics vise is a big help during this process; doing it holding the cap in your hand is just begging to stab yourself, repeatedly!):


Once the crimp is fully undone, things can be pulled out:


I used a heat gun to soften the tar holding what remained of the dried out cap in the can and then pulled it out; the softened tar can be dug out with a pick if necessary to make room for the new replacement caps:


The new cap's leads are formed and soldered to the terminal:


The negative lead is brought out and soldered to the mounting ring:



And finally the crimp is re formed by gently tapping it back down as you work around the perimeter, using a small hammer.  The cap is shown in the vise for the photo because that's where the camera was set up and focused on, but when redoing the crimp the back end is rested on the table to prevent movement from absorbing all the force of the tapping without the metal moving.  I actually rest it on the table and nest it against the heavy cast iron base of the vise, and tap at about a 45 degree angle at first to form the metal around the mounting ring and get it tight.  The blows are shifted to be more parallel to the face of the cap as the metal folds down.  Use very light hammer taps, rotating the cap slightly between each one to evenly re form the metal.  Take your time with this step and it will fold back down nicely:


The hammered part of the crimp will look a bit buggered up, but it is not visible when the cap is mounted, and I prefer this to cutting the can in a visible place.
There are a few more photos of the process at:  https://pmanning.smugmug.com/Electronics/Restuffing-can-electrolytics/51169977_mDp2vM#!i=4261375442&k=QLmNcB2

At some point when I do another I'll make a better step-by-step sequence, and do it on a multi-section cap rather than the single section one shown here.

-Pat
If it jams, force it.  If it breaks, you needed a new one anyway...
 

Offline FlyingHacker

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: us
  • You're Doing it Wrong
Re: Function Gen with Tubes - Exact 240
« Reply #28 on: October 07, 2015, 05:41:30 pm »
CubDriver...

Thank you very much that is exceedingly helpful!

Is the Cub in your name the aircraft Cub? I am a Cessna Driver myself, though I would love a Carbon Cub, or a J-3 for that matter.
--73
 

Offline Cubdriver

  • Supporter
  • ****
  • Posts: 2295
  • Country: us
  • Nixie addict
    • Photos of electronic gear
Re: Function Gen with Tubes - Exact 240
« Reply #29 on: October 07, 2015, 05:47:50 pm »
CubDriver...

Thank you very much that is exceedingly helpful!

Is the Cub in your name the aircraft Cub? I am a Cessna Driver myself, though I would love a Carbon Cub, or a J-3 for that matter.

Yes, it is, though unfortunately I haven't been driving one in far too long.  I need to get current and airborne again one of these years...

-Pat
If it jams, force it.  If it breaks, you needed a new one anyway...
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 15064
  • Country: za
Re: Function Gen with Tubes - Exact 240
« Reply #30 on: October 07, 2015, 06:10:34 pm »
I would also suggest changing out the carbon composition resitors, you can see some 47k 1W carbon composition resistors in there, where there has been a replacement of one, going from a 10% carbon composition unit to a 5% carbon film one. The composition resistors will by now have drifted, mostly high, and any value over 22k will be high by some amount, some might even be over double the printed value or even close to open circuit. Any carbon composition over 1M should be checked for resistance, simplest is in circuit with no power, and see if the measured resistance is equal to or lower than the band value, as it always will go high, so higher than the print is on the way out.
 

Offline FlyingHacker

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: us
  • You're Doing it Wrong
Re: Function Gen with Tubes - Exact 240
« Reply #31 on: October 07, 2015, 07:29:25 pm »
I would also suggest changing out the carbon composition resitors, you can see some 47k 1W carbon composition resistors in there, where there has been a replacement of one, going from a 10% carbon composition unit to a 5% carbon film one. The composition resistors will by now have drifted, mostly high, and any value over 22k will be high by some amount, some might even be over double the printed value or even close to open circuit. Any carbon composition over 1M should be checked for resistance, simplest is in circuit with no power, and see if the measured resistance is equal to or lower than the band value, as it always will go high, so higher than the print is on the way out.

I am assuming they should be replaced with new carbon composition resistors, right? I know different types of resistors have different failure modes.
--73
 

Offline TimFox

  • Frequent Contributor
  • **
  • Posts: 924
  • Country: us
  • Retired, now restoring antique test equipment
Re: Function Gen with Tubes - Exact 240
« Reply #32 on: October 07, 2015, 08:06:32 pm »
Some people think that 1960's pop music was better than later music, therefore it must have been the carbon composition resistors.  The only failure-mode difference that I am aware of is that a good A-B carbon comp resistor could take a larger transient than a carbon film, due to the larger thermal mass.
I would replace the 10% carbons with 5% or better metal films of the same wattage.  These are readily available from Mouser and DigiKey in different brands.  Note that real mil-spec metal films (e.g., RN65) have real mil-spec power ratings that are lower than the power rating on A-B carbons of the same physical size.
For example, these are $0.12 each in singles, and there are many more to be found.
http://www.mouser.com/ProductDetail/Vishay-BC-Components/SFR16S0001002JA500/?qs=sGAEpiMZZMtG0KNrPCHnjeX%2fDW687SXcfUuTu%252bLQn3c%3d
 

Offline FlyingHacker

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: us
  • You're Doing it Wrong
Re: Function Gen with Tubes - Exact 240
« Reply #33 on: October 07, 2015, 09:33:55 pm »
TimFox,

Thanks for the info. I am no expert on resistors. I know some vintage Tek experts suggested I stick with carbon composition when replacing some that were in the startup surge protection (Inrush) of the power supply on a Tek scope. I thought they said it was because carbon film can fail shorted vs. carbon composition failing open.

I doubt this is an issue here (if I even understand the issue at all).
--73
 

Offline AF6LJ

  • Supporter
  • ****
  • Posts: 2901
  • Country: us
Re: Function Gen with Tubes - Exact 240
« Reply #34 on: October 07, 2015, 10:53:04 pm »
Some people think that 1960's pop music was better than later music, therefore it must have been the carbon composition resistors.  The only failure-mode difference that I am aware of is that a good A-B carbon comp resistor could take a larger transient than a carbon film, due to the larger thermal mass.
I would replace the 10% carbons with 5% or better metal films of the same wattage.  These are readily available from Mouser and DigiKey in different brands.  Note that real mil-spec metal films (e.g., RN65) have real mil-spec power ratings that are lower than the power rating on A-B carbons of the same physical size.
For example, these are $0.12 each in singles, and there are many more to be found.
http://www.mouser.com/ProductDetail/Vishay-BC-Components/SFR16S0001002JA500/?qs=sGAEpiMZZMtG0KNrPCHnjeX%2fDW687SXcfUuTu%252bLQn3c%3d
:-+
Good advice, as a general rule I only replace the ones that are out of tolerance and leave the rest, no need for all that extra work.
Sue AF6LJ
Test Equipment Addict, And Proud Of It.
 

Offline FlyingHacker

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: us
  • You're Doing it Wrong
Re: Function Gen with Tubes - Exact 240
« Reply #35 on: October 07, 2015, 11:01:30 pm »
Good advice, as a general rule I only replace the ones that are out of tolerance and leave the rest, no need for all that extra work.

When I get the documentation I will measure the resistors to see if they are within spec. I will add them to my Mouser order for the caps.

Thanks.
--73
 

Offline AF6LJ

  • Supporter
  • ****
  • Posts: 2901
  • Country: us
Re: Function Gen with Tubes - Exact 240
« Reply #36 on: October 07, 2015, 11:41:40 pm »
Good advice, as a general rule I only replace the ones that are out of tolerance and leave the rest, no need for all that extra work.

When I get the documentation I will measure the resistors to see if they are within spec. I will add them to my Mouser order for the caps.

Thanks.
Good deal;
The carbon resistors tend to drift high especially if they have to dissipate a little power.
They also will draw moisture from the air which will lower their value slightly, not enough to cause them to drift out of spec...

Sue AF6LJ
Test Equipment Addict, And Proud Of It.
 

Offline FlyingHacker

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: us
  • You're Doing it Wrong
Re: Function Gen with Tubes - Exact 240
« Reply #37 on: October 12, 2015, 07:13:43 pm »
Hi again.

For those who are interest here is the manual for this unit (with schematic).

http://meanmutha.com/Exact_240_Function_Generator_Manual_w_Schem.pdf

Note the date is 1962 !

I see the capacitors are rated for very high voltage (500V for some). These are harder to find. But I think something like this might work for C1 and C3:

http://www.mouser.com/ProductDetail/Nichicon/LGN2H820MELZ35/?qs=sGAEpiMZZMsh%252b1woXyUXj4wVC1amv8ifvA1oLceiKgo%3d

The drawback is that they have snap-in short leads. So I would have to solder extensions on. I would probably have to solder some kind of extension on any in order to hook them up to the old leads.



C2A&B, C4A&B, and C5A&B (which I assume are cans with two caps each) would be:


http://www.mouser.com/Search/ProductDetail.aspx?R=UVZ2W220MHDvirtualkey64700000virtualkey647-UVZ2W220MHD



C9 would be:

http://www.mouser.com/ProductDetail/United-Chemi-Con/EKYA250ELL222MK25S/?qs=sGAEpiMZZMvwFf0viD3Y3dL55az7zhnmKi9Mm3rr%252bMkzd%252b7%2fxwhouw%3d%3d

I would appreciate any feedback on these choices. Thanks.
--73
 

Offline AF6LJ

  • Supporter
  • ****
  • Posts: 2901
  • Country: us
Re: Function Gen with Tubes - Exact 240 - Now with Manual and Schem
« Reply #38 on: October 12, 2015, 07:47:11 pm »
Looks like good choices to me.
Sue AF6LJ
Test Equipment Addict, And Proud Of It.
 

Offline FlyingHacker

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: us
  • You're Doing it Wrong
Re: Function Gen with Tubes - Exact 240 - Now with Manual and Schem
« Reply #39 on: October 12, 2015, 07:49:03 pm »
Looks like good choices to me.

Thanks for the sanity check. I am going to measure the resistors.
--73
 

Offline Vgkid

  • Super Contributor
  • ***
  • Posts: 2496
  • Country: us
Re: Function Gen with Tubes - Exact 240 - Now with Manual and Schem
« Reply #40 on: October 12, 2015, 09:06:20 pm »
When space is not an issue, I have put several caps in series to reach the desired voltage/capicitance.
If you own any North Hills Electronics gear, message me. L&N Fan
 

Offline FlyingHacker

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: us
  • You're Doing it Wrong
Re: Function Gen with Tubes - Exact 240 - Now with Manual and Schem
« Reply #41 on: October 12, 2015, 10:25:55 pm »
When space is not an issue, I have put several caps in series to reach the desired voltage/capicitance.

Good point. Not as important for the capacitance in this case, but, though it makes sense, I would not have thought to do that for voltage. I think I have read that in many cases this is not a good idea, though. What happens if one fails? Does the other then fail as well? Would slightly different ESRs cause the two different caps to carry different voltages?

BTW, this is an honest question. I am a hobbyist, and I don't know gobs about the subject  :scared:
« Last Edit: October 12, 2015, 10:29:05 pm by FlyingHacker »
--73
 

Offline FlyingHacker

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: us
  • You're Doing it Wrong
Re: Function Gen with Tubes - Exact 240 - Now with Manual and Schem
« Reply #42 on: October 12, 2015, 10:28:24 pm »
So I went through and measured pretty much all the resistors in circuit. A few were a tiny bit out of spec, like 11% high. It was mainly the lower values, the 10 ohms and 100 ohms that had the issues. Though a couple 1.5Mohm ones were a hint too high as well.

I think I am only going to replace the ones that are out of spec. I would rather keep the thing more authentic, and abide by the "if it ain't broke" rule.

So I am going to add these to my Mouser order and get this guy fixed up.
--73
 

Offline AF6LJ

  • Supporter
  • ****
  • Posts: 2901
  • Country: us
Re: Function Gen with Tubes - Exact 240 - Now with Manual and Schem
« Reply #43 on: October 12, 2015, 10:59:50 pm »
When space is not an issue, I have put several caps in series to reach the desired voltage/capicitance.

Good point. Not as important for the capacitance in this case, but, though it makes sense, I would not have thought to do that for voltage. I think I have read that in many cases this is not a good idea, though. What happens if one fails? Does the other then fail as well? Would slightly different ESRs cause the two different caps to carry different voltages?

BTW, this is an honest question. I am a hobbyist, and I don't know gobs about the subject  :scared:
When properly selected they don't fail, and when they do they usually open circuit.
Back in the days when transmitting tubes ruled it was common pratice to connect electrolytics in series to get the higher voltage. Equalizing resistors are used to halp compensate for differences in leakage current.

This is how you get 50 MF at 3.6KV
390MF @ 450V X 8
Sue AF6LJ
Test Equipment Addict, And Proud Of It.
 

Offline FlyingHacker

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: us
  • You're Doing it Wrong
Re: Function Gen with Tubes - Exact 240 - Now with Manual and Schem
« Reply #44 on: October 12, 2015, 11:13:12 pm »
Interesting... The resistors are a good idea. Thanks.
--73
 

Offline AF6LJ

  • Supporter
  • ****
  • Posts: 2901
  • Country: us
Re: Function Gen with Tubes - Exact 240 - Now with Manual and Schem
« Reply #45 on: October 12, 2015, 11:18:36 pm »
Interesting... The resistors are a good idea. Thanks.

When you series connect caps you need them, generally speaking the equalizing resistors should be of a value to equal 5 to 10 times the leakage current of the caps.
General rule of thumb anything in the neighborhood 100K ohms is good for voltages in the 600V to 4KV range. You do need to be careful of one thing; The resistors must be rated to hold off the voltage they are going to be subjected to. Those resistors are rated for 800V.

You are really better off if you don't have to do this.
When I rebuilt that power supply I used good caps (Panasonic) that had a reasonablely high ripple current rating since they were used in a voltage doubler.
« Last Edit: October 12, 2015, 11:20:37 pm by AF6LJ »
Sue AF6LJ
Test Equipment Addict, And Proud Of It.
 

Offline FlyingHacker

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: us
  • You're Doing it Wrong
Re: Function Gen with Tubes - Exact 240 - Now with Manual and Schem
« Reply #46 on: October 14, 2015, 01:13:39 am »
I placed my parts order at Mouser. I got all the caps I listed above, which were all the proper voltage rating. For those playing along I also made sure my resistors had a high enough voltage rating. Many do not. I just got them all for 700V. They were not particularly expensive, and I wasn't buying too many of them (only bought 100 if the price break was decent enough).

Looking forward to recapping the recap.... :-DD
--73
 

Offline AF6LJ

  • Supporter
  • ****
  • Posts: 2901
  • Country: us
Re: Function Gen with Tubes - Exact 240 - Now with Manual and Schem
« Reply #47 on: October 14, 2015, 01:31:17 am »
I placed my parts order at Mouser. I got all the caps I listed above, which were all the proper voltage rating. For those playing along I also made sure my resistors had a high enough voltage rating. Many do not. I just got them all for 700V. They were not particularly expensive, and I wasn't buying too many of them (only bought 100 if the price break was decent enough).

Looking forward to recapping the recap.... :-DD

Good deal, That is going to be a fine generator when you get finished.
Sue AF6LJ
Test Equipment Addict, And Proud Of It.
 

Offline FlyingHacker

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: us
  • You're Doing it Wrong
Re: Function Gen with Tubes - Exact 240 - Now with Manual and Schem
« Reply #48 on: October 20, 2015, 09:45:51 pm »
Wow! Actually restuffing the metal cans is a massive pain. Since this unit has cardboard tubes, and no metal cans exposed I am thinking about just putting the new caps in the cardboard, rather than trying to restuff all the metal cans.

I don't have a heat gun, and it seems we no longer have a hair dryer either. I got into the can, but I couldn't get the inside plastic cap off without heat (or fear of breaking it). On the first one I did I managed to pull the contacts off the tabs that go into the black plastic inside cap.

Any reason not to do it this way? It will look the same. I will have to glue the cardboard to the phenolic wafers somehow. Not sure what the best glue to use would be. If I fill them with hot melt it would work, but that would seem to ruin the tubes if I needed them later.
« Last Edit: October 20, 2015, 09:51:43 pm by FlyingHacker »
--73
 

Offline AF6LJ

  • Supporter
  • ****
  • Posts: 2901
  • Country: us
Re: Function Gen with Tubes - Exact 240 - Now with Manual and Schem
« Reply #49 on: October 20, 2015, 11:51:15 pm »
You could use silicone glue (RTV or equiv.) to glue down the cardboard tubes.
Anyone have better ideas??  8) :)
Sue AF6LJ
Test Equipment Addict, And Proud Of It.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf