开云体育

On Thu, Mar 3, 2022 at 12:41 PM, Jim Ford wrote:

Ah, the old plastic transparencies!? The late, great Bob Pease of National
Semiconductor used to refer to them as "analog PowerPoint".

He used that phrase on the Bob Pease show (of which far too few episodes exist, at least on YouTube), but I thought he was referring to acetate sheets, not vinyl sheets.

-- Jeff Dutky

On Tue, Feb 1, 2022 at 05:29 AM, Colin Herbert wrote:

According to the manual, C26 is supposed to be an Aluminium electrolytic at
50uF +100% -10% and 6 vdcw, whereas C27 and C28 are also Aluminium
electrolytics at 50uF +75uF - 10uF and 25 vdcw. All three were originally made
by Sprague. I realise that these would now be quite old (the manual was
printed in October 1975) but what would one use to replace them? Instinct
suggests that modern capacitors would probably better that those used in 1975,
but 50uF is a value that doesn't seem to exist these days.

Colin -

Generally if the above are radial power supply caps with white or silver-ish look, likely after 50 years, they are bad. Any modern ones will be physically smaller, and likely have a somewhat lower capacitance in reality (like -20% from what's printed on the outside) so I'd fit a 220uF or 330uF/10 V for the 50uF/6 V; for the 50uF at 25V, I'd use a 330 or 470uF at 25 or 35 Volts, usually you can't go wrong when going up in value five to ten times, especially since vintage caps, being physically larger, usually had lower ESR (when new, but no longer) than modern (physically smaller) caps. Since the set might have tantalum caps, check for those, the 1970-1975-ish parts can fail over time, might be smart to replace them while the set is already apart.

Steven

It may not matter for this attenuator.? Definitely it does on some high power models.? I've seen fixed attenuators that can handle hundreds of watts into one port but 50 W or less into the other.? The manufacturer saved money by only beefing up the elements at one end.Jim Ford?Sent from my T-Mobile 4G LTE Device

Hi All,

I have one of these older Analyzer w/Mainframe power supply units (1980-84?) on its way to me.

Any known failure modes?

Anyone ever refurbish / Improve / Upgrade one?

Does it have old failure-prone Tantalums?

Is the PS likely to need Lytic caps?

Thanks!



Steven

I have a JFW 50DR-077, this attenuator is cylindrical with rotating dials on the front.

Does anyone know which port is the input and which is the output, it doesn't seem to be labeled so I can't tell

Sounds like a monumental achievement!
As for the 2N2919, I have a couple to spare if you can’t find a good price on enay.

Can you obtain Poly-Aluminum capacitors? Those are my choice when replacing Tantalum capacitors.

-Chris

Ah, the old plastic transparencies!? The late, great Bob Pease of National Semiconductor used to refer to them as "analog PowerPoint".Jim Ford?Sent from my T-Mobile 4G LTE Device

Or a dual transistor in one package?? That would be about the best they could be matched.?Jim Ford?Sent from my T-Mobile 4G LTE Device

Morris,

You will not do any harm by using two separate transistors but depending how well the two Vbe are matched you will get an output which is not precisely +15V and since you have only one adjustment for all the low voltages you will not get them all spot on at the same time. You also have the differential drift with temperature which means the +15V will change as the scope warms up.

How much this matters in practice depends on how well matched are your discrete devices, how well you clamp them together for thermal matching and how tolerant you are on changes in calibration. I would be very tempted to match up a pair from a batch of 10 or so, clamp them with some brass or aluminium strip and just try it out. Costs very little and you may find you have an adequate solution for your needs.

Roger

On Thu, Mar 3, 2022 at 04:02 PM, Albert Otten wrote:

Interleaved.

Stephen,

If the voltages (at the EL84 pins) are still exactly the same, do you mean as
in a previous message? Let me recall:
2- Pin 6: +3V
3- Pin 7: +1.6V
4- Pin 8: -4.7V instead of -145V
5- Between Pin 2 and R55: +3V instead of -151V
At pins 6 and 7 you measure Vg2 and Va. How can these ever be positive? Did
you perhaps not measure w.r.t. GND?

Yes I definitely did measured referenced to chassis ground.

With my 106 at minimum amplitude I get Vk
May I ask again whether or not Vk (= "A") can be varied with the Amplitude
knob to at least -150 V or so?

I don’t think so, but I’ll check again later.

What I can tell you, however, is that when I connect a scope to the Fast-Rise output, the flat signal (AC coupled) goes down when the intensity knob is turned up, and up when the intensity knob is turned down…

About pin 6 just to be sure: are pins 6 and 9 internally connected in the
EL84? See also a remark by David who thinks no modifications are needed.
Philips labels "i.c." at pin 6, which means do NOT connect this pin to
anything.

Albert

I understand what David is saying, and I’m also going to respond to him here.
David, all of the EL84 pinouts I’ve checked refer to pins 1, 6 and 8 to be internal connections.

However, what greatly confuses me is that the unit was perfectly working before WITH pin 9 entirely disconnected, and pin 6 connected. The unit had no modifications whatsoever, as if it were running with 7189A’s without modifications…. I don’t get it.

But I will do the modification before I do anything else.

I will report back as soon as I can.

Dave,

That (making an overlay) was exactly the bush that I was beating around (or building up to?). It looks like you can buy clear vinyl sheet that can be printed on with a laser or ink jet printer, so you could have an overlay that would adhere to the implosion shield like a Colorforms sticker (search for Vinyl Sticker Paper on Amazon).

I'd try drawing up your own. I'm not sure that Word or PowerPoint are suitable; maybe Visio would be a better tool? Better still would be an actual drafting program. I don't know what there is for Windows, but on the Mac there is an old and reliable program called MacDraft by a company called Microspot. They have a free 14-day trial version that should be more than adequate.

I would expect that even without a good image you can calculate where the vertical lines are supposed to go based on measurement of the existing CRT graticule. The vertical lines clearly don't line up with the graticule lines, but each line is clearly labeled, and the left and right edges might be aligned to the left and right edges of the graticule.

-- Jeff Dutky

I was wrong! You absolutely do need to modify the socket wiring to use EL84.
Some EL84 datasheets show pin 6 as IC instead of NC.
Tek wired the V64/74/84/94 screen grids using both pins 6 and 9. If 6 in your tube goes to anything but G2, it won’t work to say the least.
Take the wires going to pin 6, and connect them to pin 9 instead.

Sorry,
Dave Wise

From: [email protected] <[email protected]> On Behalf Of Stephen via groups.io
Sent: Thursday, March 03, 2022 5:38 AM
To: [email protected]
Subject: Re: [TekScopes] Type 106 Saga (Again)

David, Albert,

I rechecked the voltages, this time with the correct settings, and they are still exactly the same.
I also checked all the waveforms to be expected on the “GENERATOR” schematic.
All waveforms are perfect until I get to R99 (600Ω). From this point and everything beyond (to the right), zilch. Flat as a pancake. Which was to be expected since there is no output.
Q54 is fine. For good measure I replaced it temporarily. Same results.

Although the TP voltages are all fine, I’m seriously considering recapping the power supply… again.
At this point, I don’t know what else to do.

What do you think?

Sphere has their "Hoods, Filters, & Bezels" page, , but doesn't include the 5000 series audio log scale screens. Given what I've seen they are actual implosion screens, not overlay transparencies.

Walter, despite not having any available (didn't expect you would), do you know the part number?

I've seen one on a scope on eBay before, and was quite tempted to attempt to buy it just for the screen. I think it might have been the last 5L4N equipped scope that was up for auction and was snipped by an auto-bidder. So wouldn't have gotten it anyway. And I think it'll be vastly more economical to make my own than deal with purchase of such a rarity.

My grand plan is to print a properly sized overlay on clear transparency that I can layer on a clear implosion shield. This place is full of old guys like myself? We all remember the days when we printed presentation materials on clear 8 1/2" x 11" transparencies, right? Foils, we used to call them. Heck, we were still using those when I started at Intel in '91. I think I've seen boxes of them stuffed in drawers in the printer rooms at work. I think all the printer rooms were cleaned out during COVID shut-down. But next time I'm in the office I'm going to rifle the drawers for them.

My hope is someone out there has a PDF with an overlay ready to go, or knows where Tek might have such a thing already. The only thing I've found is the example at the end of the audio cookbook. I did a test print of that last night. It would take a fair bit of fussing to get scaled properly. I might be better off building my own in PowerPoint or Word.

The EL84 datasheet I looked at listed pin 6 as NC. Safest to do the mod.

Dave Wise

From: [email protected] <[email protected]> On Behalf Of Albert Otten via groups.io
Sent: Thursday, March 03, 2022 7:02 AM
To: [email protected]
Subject: Re: [TekScopes] Type 106 Saga (Again)

Stephen,

If the voltages (at the EL84 pins) are still exactly the same, do you mean as in a previous message? Let me recall:
2- Pin 6: +3V
3- Pin 7: +1.6V
4- Pin 8: -4.7V instead of -145V
5- Between Pin 2 and R55: +3V instead of -151V
At pins 6 and 7 you measure Vg2 and Va. How can these ever be positive? Did you perhaps not measure w.r.t. GND? With my 106 at minimum amplitude I get Vk = -6.4 V, Vg2 = 0 V, Vg1 = SW roughly between -7 V and -22 V, output SW between 0 V and -0.2 V (at 50R) or -2 V (unterminated).
5) Between Pin 2 and (top of) R55 is the voltage drop across R55 and equals -Vgk, not that -151 V. The average value of Vg in my case was about -15 V, so Vgk about -8.5 V on average.
May I ask again whether or not Vk (= "A") can be varied with the Amplitude knob to at least -150 V or so?

About pin 6 just to be sure: are pins 6 and 9 internally connected in the EL84? See also a remark by David who thinks no modifications are needed. Philips labels "i.c." at pin 6, which means do NOT connect this pin to anything.

Albert


On Thu, Mar 3, 2022 at 02:38 PM, Stephen wrote:

David, Albert,

I rechecked the voltages, this time with the correct settings, and they are
still exactly the same.
I also checked all the waveforms to be expected on the “GENERATOR”
schematic.
All waveforms are perfect until I get to R99 (600Ω). From this point and
everything beyond (to the right), zilch. Flat as a pancake. Which was to be
expected since there is no output.
Q54 is fine. For good measure I replaced it temporarily. Same results.

Although the TP voltages are all fine, I’m seriously considering recapping
the power supply… again.
At this point, I don’t know what else to do.

What do you think?

Dave,

Are the overlays replacement plastic windows for the CRT bezel, or are they something that goes on top of the plastic window?

How precise do the overlays need to be? I assume that the X-Y translation of the overlay is not important, since you can easily shift the trace on the CRT horizontally and vertically, but how precise does the spacing between the vertical marks need to be?

I thought that I saw something like these on Sphere's stuff page last year, but those may have been for a 500-series scope, or maybe something completely different, and I certainly have not seen them listed recently.

-- Jeff Dutky

Stephen,

If the voltages (at the EL84 pins) are still exactly the same, do you mean as in a previous message? Let me recall:
2- Pin 6: +3V
3- Pin 7: +1.6V
4- Pin 8: -4.7V instead of -145V
5- Between Pin 2 and R55: +3V instead of -151V
At pins 6 and 7 you measure Vg2 and Va. How can these ever be positive? Did you perhaps not measure w.r.t. GND? With my 106 at minimum amplitude I get Vk = -6.4 V, Vg2 = 0 V, Vg1 = SW roughly between -7 V and -22 V, output SW between 0 V and -0.2 V (at 50R) or -2 V (unterminated).
5) Between Pin 2 and (top of) R55 is the voltage drop across R55 and equals -Vgk, not that -151 V. The average value of Vg in my case was about -15 V, so Vgk about -8.5 V on average.
May I ask again whether or not Vk (= "A") can be varied with the Amplitude knob to at least -150 V or so?

About pin 6 just to be sure: are pins 6 and 9 internally connected in the EL84? See also a remark by David who thinks no modifications are needed. Philips labels "i.c." at pin 6, which means do NOT connect this pin to anything.

Albert

David, Albert,

I rechecked the voltages, this time with the correct settings, and they are still exactly the same.
I also checked all the waveforms to be expected on the “GENERATOR” schematic.
All waveforms are perfect until I get to R99 (600Ω). From this point and everything beyond (to the right), zilch. Flat as a pancake. Which was to be expected since there is no output.
Q54 is fine. For good measure I replaced it temporarily. Same results.

Although the TP voltages are all fine, I’m seriously considering recapping the power supply… again.
At this point, I don’t know what else to do.

What do you think?

Hi Mark,

The aim with this scope is to identify the problems and repair with whatever is available here, or at least until I can buy the correct components when next in Europe. I say that, as tantalums (like low ESR) are almost non-existent over here, so the scope will get whatever is available for now. Mouser only supply from abroad and apart from Brazilian pay rates making Mouser impossible (I'm often lucky if I make $30 a week), the postal service here cannot be relied upon to deliver. I know Brazil is sunny and appears exotic, however it is more like Nigeria in terms of efficiency and corruption.

Anyway, all the transistors on the A6 have been tested although there is an anomoly on the adjacent A10 board, where the appears to be a near short (14Ω) across two caps on the 15v rail and also a near short across all the 74xxx series chips and the 5v rail at 9Ω.

Have to head out for the day but shall investigate further this evening.

Hi all,

I've finally got the 7904 power supply to stop ticking and produce all its voltages properly - what a job! This one worked briefly when I got it, just long enough to show me that one of the two 7904s I have has a dud CRT. When I transplanted it into the other scope it started ticking and led me a merry chase. I had to make a test load box to run it outside the scope - the box includes a voltmeter and LED indicators for each voltage. As anyone who has had to work on one of these supplies will know, the power for the control IC is derived from a current transformer in series with the primary of the 25 KHz power transformer so you need a minimum load for the IC to be powered properly and get the supply to work. For anyone going down that path, here are the loads I used (courtesy of the junque box):

+130 .. 2.2K 10 watts
+ and -50...250 ohms 20 watts each
+ and -15...82 ohms 10 watts each (I would have used lower values, perhaps 33 or 27 ohms 10 watts if I had them )
+5...10 ohms (probably could use 4.7 ohms)

The total load is therefore 35.6 watts.

The faults I found in this supply were:

Two shorted diodes in the +5 volt lamp supply - replaced with 40 volt 3 amp Schottky diodes.
Several tantalum caps with high ESR but none shorted - many replaced anyway.
A faulty dual transistor in the +15 supply. This is specified as a a 2N2919 or NS7348. I was able to use one from the other dead supply but it would be nice to have a replacement. They are available at a high price but I was wondering whether it would be possible to substitute a pair of reasonably well matched discrete transistors. They are used as differential pairs and there's lots of feedback around the circuits that should minimize the need for precise matching. Has anyone tried that?

Anyway it's late here and it's been a busy day which featured a "long lunch" so I'll put it all together and into the 7904 in a day or two and report back.

Cheers,

Morris