Adding Video Trigger to TDS744A

[rochoz]

No problem. Let us know how things turn out.

Rochoz

[hkp7]

Just one additional thought; is it possible to download the NVRAM contents via the GPIB? It would seem that using the EPROM programmer would be more direct and probably easier. Maybe I just answered my own question.

[rochoz]

I really wish there was a way to do that, I know of no way to do it. Doing this would eliminate a lot of the risk in the desoldering approach.

I have actually tried doing a similar approach by cabling my EPROM programmer to the scope using SOIC test clips on the address and data bus and holding the CPU in reset. Unfortunately, this didn't work because the buffers on the EPROM programmer aren't strong enough to drive the address and data bus (I could see the signals were poor on a scope). I've thought about designing a small board with 245 transceivers to buffer the address and data, but haven't gotten that motivated yet. I have found the necessary enable pins to tri-state the address and data signals, so in theory it should work.

Reading it from the GPIB would be a very cool setup, but I don't know how to do this.

[hkp7]

If I can figure out a way after I'm done, I will post it. I had read on some posts not to replace the old caps with electrolytics and on others saying use good quality aluminum electrolytics and no substitutions. What is your take on that? I was about to order caps and have been trying to decide what to use; 7000 hr/105C aluminum electrolytic, solid electrolytic, or tantalums. I was leaning towards the 7000 hr/105C aluminum caps. Any opinion?

[rochoz]

My opinion is the long life/high temp electrolytics is the way to go. I haven't been all that picky actually with a certain life/temp rating, I usually try to find a good combination of the long life and higher temp at a reasonable price, because there's a ton of them on the 5XX and 6XX series scopes.

As I recall the TDS744A doesn't have very many electrolytics, does it? From what I can recall, the early vintage 540/640/540A/640A series of scopes had a ton of them, and as the scopes got newer the use of the electrolytics became less and less. Since the 744A is a bit newer maybe I thought maybe there weren't that many electrolytics....but I don't have one to look at so I don't know.

[hkp7]

Mine is a 544A. As far as I know, the 744 is supposed to be without the problems of its earlier cousins. I was leaning towards the good quality electrolytics but had that little bit of doubt about using alternatives. Since you've had successes in the past, I value your opinion because it's not based on guesswork. I will go ahead with the electrolytics. In my mind, since they will be soldered in place using an iron and not reflowed, they wont be subjected to the excessive heat that caused the originals to fail to begin with.

[rochoz]

Ahh, okay, I was looking at the subject of this thread and assumed you were talking about a TDS744A.

Yes, the 544A has a ton of caps on both the Aquisition board as well as the CPU board. I've used Panasonic AHC and VHC series caps from Digikey, and I've also used Nichicon UCD series caps from Mouser with no problems. Like I say, I usually use the price of the caps as a guage for which caps to buy. I don't think the original caps that were used on these scopes were either high temp, long life, or any other special characteristics. So in my opinion, just about any cap is probably better than the original ones.

The other thing to keep in mind is that at the time these scopes were first designed and manufactured, SMT electrolytics were just starting to enter the scene. My guess is these caps had problems with their reliability, and nobody knew about it until years later when they discovered that these things leak their electrolyte. And by that time it was too late to make any changes as the units were out of warranty and/or out of support, and Tek was already well into designing the next series of scopes that didn't use the electrolytics.

The only thing I would consider replacing the electroltyics with would be a ceramic cap. I've looked into that, but you'll spend a small fortune buying ceramic caps of the same value as the electrolytics that are on there....and I'm not convinced that the cost would be worth it. In my opinion, using decent quality electrolytics as a replacement will likely last a very long time. Some people may disagree with me here, but that's my opinion for what it's worth.

One other thing. The front panel board also has electrolytics that should be replaced, I think around 6 of them. But 2 of the caps are actually the wrong voltage for the circuit!! There are 2 of a 33uF/10V cap used on the +15V and -15V rails. This is a design problem, and you should replace those caps with a higher voltage rated unit. I used a 33uF/25V cap in the same case size for those caps.

Good luck.

[hkp7]

I was considering the nichicon UCB series they are less than $0.30 each from mouser, spec sheet http://products.nichicon.co.jp/en/pdf/XJA043/e-cb.pdf

[rochoz]

Those should be fine, but from what I can see the 33uF/10V parts is a non-stock item for Mouser. For reference, the UCD series caps are 0.16 each. The UCD series is not necessary "long life", but they are rated for 5000 hrs at 105C, and they have the low impedance characteristic.

[hkp7]

That is true; from what I can tell, none of them are. I don't know what the lead time is from them either. In any case. I was planning to use caps with a higher rating than the original 10V figuring that the derating should contribute to longevity maybe overkill? The only reason I was focusing on the nichicon was because I've read in other posts that they have a reputation for never leaking (electrolyte). I was also considering the Panasonics despite the fact that they were the suppliers of the originals. I think all of the major component suppliers have learned from the past as far as susceptability to failure goes.

I got the universal programmer today and played around with it on a DS1250. It seems all I need to do is read the chip, put the info into the buffer and then save it. Does the information get saved as hex or bin? I'm thinking hex since bin would probably be used to make a CD ROM. The English translation of the manual definitely leaves a lot to be desired!