Author Topic: Attempting to recreate a Tektronix Test Card (Read 2236 times)

Zenwizard · « **on:** July 19, 2021, 06:05:30 pm »

I am diving in to the deep end of the pool and seeing where this goes. I have a Tektronix 7854 that I would like to do a full calibration on so I can use it on the bench. But there seems to be a very illusive card that I need for the calibration. A 067-0912-00 Test card. This card allows for the 7854 to be operated without the digital sections installed in the frame. But it allows all the analog functions to work as well as channel switching and specifically the vertical amplify thermal compensation. To be adjusted. If I can get a test card or make a replacement test card I am considering a re-cap of the 7854 to make is last the long term.

I have run in to a snag. The test card uses a 74LS314 1024-Bit Random Access Memory (RAM) ic. Is there a modem version of this chip or am I dead in the water? I am very familiar with digital logic but I have never really had to use ram before so in uncharted territory at least for me.

Attached in the only data I have been able to locate for this card.

Thanks in advance for the advice.

radix · « **Reply #1 on:** July 19, 2021, 07:37:42 pm »

That's a 7, not 1, so those are 74LS374, which you should have no trouble sourcing.

They're also listed in the part list as an octal D FF and you can also look up the Tektronix part number in the Tektronix Semiconductor Cross Reference when in doubt.

Zenwizard · « **Reply #2 on:** July 19, 2021, 09:37:48 pm »

Well... I missed that one. I will continue to work on the card now Thanks.

Renate · « **Reply #3 on:** July 20, 2021, 10:47:15 am »

That's a pretty simple button latch.
The pullups on the right are only necessary if you want to do the "all bulbs on" test with jumper P430.
(You can test them individually by just hitting the eight buttons.)

Me? I'd probably just use an Atmel ATMega328P µC.
One chip and you're done.
(And you could make it do chasing lights too.)

Zenwizard · « **Reply #4 on:** July 20, 2021, 02:19:33 pm »

Renate a micro would work however There is something else that the board does that is very critical other then test the lights. If all it did was test the lights I would not really care due to mine already being replaced with LED’s. I found soft white to be a good color for the scope.

The other function of the board Is to control the analog circuitry without the digital board in place. This allows the chassis to be run as a analog unit with out any of the processing. The test board is used only in this scope so they are super rare in the wild. It also required analog terminations for several signal lines to keep everything will behaved.

Zenwizard · « **Reply #5 on:** July 20, 2021, 04:24:12 pm »

Need to find a modern replacement for a 2N5769. This is labeled as a "switching transistor" but the data sheets don't say anything about switching frequency. However the voltage and currents look very modest on it. Any one know of a good replacement sence this is a redesign of the board package really does not matter. As I can make the footprint to suite.

Renate · « **Reply #6 on:** July 20, 2021, 05:50:02 pm »

Quote from: Zenwizard on July 20, 2021, 02:19:33 pm

Renate a micro would work however there is something else that the board does that is very critical other than test the lights.

Yes, it makes the buttons functional.
But all that stuff with inverters, latch, pullup, NANDs, strobe could be done with far few connections in this day and age.
8 buttons in, 8 data lines out, one strobe line. EZ-PZ.

Quote from: Zenwizard on July 20, 2021, 04:24:12 pm

Need to find a modern replacement for a 2N5769. This is labeled as a "switching transistor" but the data sheets don't say anything about switching frequency.

Doesn't the manual for this board says you are using the probe calibration signal?
What's that, 1 kHz?

Zenwizard · « **Reply #7 on:** July 20, 2021, 05:57:44 pm »

It is true the card is tested at 1Khz but I think this is part of the trigger pass thru when it is actually in use. In that case it needs to go up to 400 Mhz and beyond. But as this is a cal card the answer I feel should be as fast as possible with in reason. If that is the case I would need the fastest transistor that I could find Vce > 15 V with a Hfe of about 140. Should not be too awful to find given modern manufacturing and with package not being an issue. Ill be able to do a dive in to data sheets when I get back to the lab this evening.

Zenwizard · « **Reply #8 on:** July 21, 2021, 12:02:18 am »

**Update**
The transistor in questions is a 500 Mhz transistor so it seems we do need to go fast here. I have located a sutable replacement I think. At the moment my current target is a

SMMBTH10-4LT3G
https://www.mouser.com/ProductDetail/ON-Semiconductor/SMMBTH10-4LT3G?qs=abmNkq9no6AVRlwMthL29A%3D%3D

Vce - 25Vdc
Hfe - 120 -> 140
Switching frequancy 800 Mhz

Package
SOT-23

David Hess · « **Reply #9 on:** July 21, 2021, 01:44:14 am »

The 2N5769 is a fast saturated switch which means that it has a low delay time. It cannot be replaced with any common RF transistor unless the design is changed to use Baker clamping which is not always practical.

Similar available NPN parts are the MMBT2369 and BSV52.

bdunham7 · « **Reply #10 on:** July 21, 2021, 02:19:14 am »

Quote from: Zenwizard on July 20, 2021, 04:24:12 pm

Need to find a modern replacement for a 2N5769. This is labeled as a "switching transistor" but the data sheets don't say anything about switching frequency. However the voltage and currents look very modest on it. Any one know of a good replacement sence this is a redesign of the board package really does not matter. As I can make the footprint to suite.

Quest has a few of the originals left. Look on Octopart. You'll pay a premium, but the fewer parts you substitute the greater chance of ending up with a working unit. Otherwise you need to look at this part of the datasheet.

Zenwizard · « **Reply #11 on:** July 23, 2021, 07:03:21 pm »

Here is the schematic mostly drawn up. I need to work on making Ki cad do a card edge connector well 2. This is proving the be a bit more challenging that i expected but it should be just fighting with software. But forward progress is being made.

Renate · « **Reply #12 on:** July 23, 2021, 07:57:27 pm »

Well, if you don't mind fighting with software, why don't use do the ATMega328P?
It will do what you want.
That's one IC instead of seven.
It's a lot easier layout.

Zenwizard · « **Reply #13 on:** July 28, 2021, 12:57:27 am »

Well I got what the connectors in the scope need to be.

The Large connector in the front of the scope is a AMP 3-53-0662-0-8301

The smaller connector in the back of the scope is a AMP 2-530662-5-8110

So making slow progress

Zenwizard · « **Reply #14 on:** August 01, 2021, 05:13:23 pm »

Here is the most resent version of the schematic. Sorry it is a little chaotic. This has been interesting to try and make sense of. I have to finish getting the test points on the schematic but after that I should be on to layout. The layout will not be as chaotic for sure.

Zenwizard · « **Reply #15 on:** August 27, 2021, 03:09:04 am »

OK I think I am at the point where I might have a prototype card for the Tektronix 7854. This is the most ambitious layout I have tried to date so feed back is extremely helpful I know there is a lot going on on this board and it is rather large for what it is but I was shooting for duplication of functionality. There is some jelly bean TTL logic on the board but that is for testing lamps and getting the scope in an operable state with out any of the digital boards installed. There is the high speed switching portion that is the impotent part for the calibration to toggle the scope for the thermal compensation of the vertical section. If there is nothing glaring the next step will be spinning some boards and assembling some prototypes.

Zen

Renate · « **Reply #16 on:** August 27, 2021, 12:36:03 pm »

Mmm, I see a lot of little problems.
First, there's no need to make the traces so small, you've got plenty of room.
It might make no difference here, but don't use 90° bends, use more 45°.
There's a few weird snap problems where you have weird squiggles to make things meet.
Some of the traces on the same layer are overlapped.
You might have better luck if you move all the pullup resistors from the far side of the ICs to between the connector and the ICs.
You're missing a bunch of vias to connect green traces to the edge connector.

Zenwizard · « **Reply #17 on:** August 27, 2021, 02:16:33 pm »

Thanks for the feed back. I will be spending more time in the digital section over the weekend. The edge connector is double sided. the green traces need to be on the lower level of the board.

Kleinstein · « **Reply #18 on:** August 27, 2021, 02:37:51 pm »

I would add decoupling caps (e.g. 100 nF X7R) for the logic chips. It is better to at least have the option to have them.
THe original plan mentions bypassing too, though not drawn.

At least the ground traces should be thicker. This also helps to directly recognize the ground.

Zenwizard · « **Reply #19 on:** September 16, 2021, 06:00:02 pm »

I have iteration 3 layed out. If the layout looks good I plan on adding some small bypass capacitors to the IC's and do a ground copper pour. Then I will send it off to the fab.

Zenwizard · « **Reply #20 on:** September 20, 2021, 04:01:53 am »

Rev 4 is complete and I hope this will be a final that I can send off to the fab. The 2 PDF's are with ground planes and with out. I think these are ready for real world prototyping as I have done all I can do in the virtual world.

Zenwizard · « **Reply #21 on:** September 20, 2021, 07:22:16 pm »

A few more last minute revisions


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: Attempting to recreate a Tektronix Test Card (Read 2236 times)

Share me