• Please review our updated Terms and Rules here

PET 2001 Booting Issues

ryanhayward said:
Hey All, So I am thinking I should invest in a good cheap Oscilloscope
Click to expand...
Buy a logic probe, should be under $20

>>> FLAMEPROOF_UNDERWEAR ON

Don't buy a crappy old analogue scope... just don't do it.

If you have budget buy a 4CH Rigol DS1054Z

If you don't have budget then stick with the logic probe and possibly add a cheapo clone 8CH logic analyser from ebay
e.g. https://www.ebay.co.uk/itm/363813342090
The trouble I have with logic analysers is the wires fall off... so you might need an IC clip too..
 
While analogue scopes are more satisfying to use and excel in certain areas, I think the DS1054Z (or similar) is the more useful tool when dealing with these kind of faults.
Nivag Swerdna said:
Don't buy a crappy old analogue scope... just don't do it.
Click to expand...
A good analogue scope is a nice addition when you can get it for a really good price. I agree to not buy a broken / crappy one as the main scope though.
 
ryanhayward said:
What are you thoughts on the https://www.amazon.ca/Hantek-Digital-Storage-Oscilloscope-Bandwidth/dp/B009H4AYII/ref=sr_1_15?crid=JZVM906WO3TU&keywords=oscilloscope&qid=1655390572&sprefix=osc,aps,85&sr=8-15#customerReviews Hantek? it is 20mhz, Adrian Black did a review on them recently and said they where OK for most use, I am also thinking I can use this as part of my Computer Science course next year and the ability to display it on a computer screen to cast to a tv is handy, thoughts?
Click to expand...

The Hantek does seem to get reasonable reviews. People say that there's a fairly steep learning curve when learning to drive them. As a long time analogue scope user I'm not best placed to advise but I can see the attraction of a piece of kit that could be put to use in other settings.

Alan
 
ajgriff said:
The Hantek does seem to get reasonable reviews. People say that there's a fairly steep learning curve when learning to drive them. As a long time analogue scope user I'm not best placed to advise but I can see the attraction of a piece of kit that could be put to use in other settings.

Alan
Click to expand...

Like all else when it comes to test instruments in the laboratory "The best instrument in the laboratory is the thing between the ears".

There are all sorts of arguments about whether a digital or a an analog scope is better. A straight scope or a storage scope too. None are particularly valid on their own, for comparison purposes because it is like trying to compare apples & pears most of the time.

You primarily need to consider what you want the scope for.

Test and measurement, data acquisition & mathematical waveform analysis, calibration applications, and USB interconnectivity ..all where memory can come in handy, calibrated cursors etc there are a myriad of functions that most simple analog scopes cannot do and a Digital scope is the go. Though the high end analog scopes like the 2465B have a CPU & firmware, have memory and calibrated cursors for amplitude, frequency and period as well as a dual timebase , so with that you almost get the best of both worlds, but you require a separate storage scope for very slow infrequent events.

Or is it for waveform display and simple non precision amplitude & frequency measurement from the scopes graticule that you want , because what you need is a useful tool to show your brain quickly and reliably the waveform on a pin or circuit node to help you with repairs ?

If this latter situation is the case, it is impossible to beat the Analog scope. The ability of these scopes to free run and be easily synchronized and the character of the artifact free display makes them unbeatable. I once bought a digital scope and found it could not do a free running trace and require a solid trigger signal, it was "hopeless".

One interesting thing about an anaolg scope display too, there is more information on the screen than meets the eye. This is because the CRT's beam brightness is inversely proportional to its velocity, So fast rise parts of a wave are less bright than periods where the voltage is stationary or has a lower rate of change. A digital scope is usually a "line drawer" and essentially draws an outline of the wave with the same thickness row of pixels over all parts of it, often with steps that are not really there due to the quantization of the the original signal. And there can also be discontinuities in it.

In a nutshell, for fault finding with the analog scope, what you see is what is there, there aint nothin more to it (to sort of quote Tina Turner). With a little imagination, you could think of the display you see on the Analog scope as being constructed from an imaginary super digital scope which had a sample interval of the Planck time of 5.39 × 10^-44 seconds, free from any artifacts of a digital design, and where the designer made the display waveform thickness & brightness inversely proportional the waveform's rate of change.
 
Last edited:
Hugo Holden said:
If this latter situation is the case, it is impossible to beat the Analog scope. The ability of these scopes to free run and be easily synchronized and the character of the artifact free display makes them unbeatable. I once bought a digital scope and found it could not do a free running trace and require a solid trigger signal, it was "hopeless".
Click to expand...

What kind of DSO was this? I haven't used a DSO in the last 25 years with a timebase that couldn't run "free". Just put the timebase in "Auto" mode and wind the trigger-level cursor off the screen. And I'd like to see an analog 'scope with a usable "roll" mode. Being a bit of an anlogue enthusiast, I have about 20 oscilloscopes and still resort to the analog occasionally, often because my 200 MHz / 2GSa/s Rigol DS1202CA can't match the bandwidth of my Tek 7904A.

The Rigol is over a decade old now and it has a pathetic memory depth which means it can't do things like properly display an AM or SSB modulation envelope, so nearly any of my analogue 'scopes with at least 20 MHz bandwidth are better for that task. However, all but the crappiest low-end DSOs these days have plenty of memory depth, allowing them to maintain high sampling rates at low timebase settings, so aliasing really isn't much of a thing with DSOs anymore. Oh, and the "digital phosphor" concept (trademarked by Tektronix and copied by nearly everyone else) of trace intensity modulation has got be be nigh on two decades old now. It was a bit crap back around about the year 2003 or 2004 when I first got the opportunity to asses a DSO with the capability, but things have come a long way since then, even for the budget machines. Skip to 6m 30s in the video here:

1655514547465.png


But this is a vintage computer hardware forum, not a HAM radio forum. While nearly any old analog 'scope is better than no 'scope at all, when it comes to digital stuff, the single-shot timebase and storage capabilities of a DSO, even one as old as my Rigol DS1202CA, renders an old analog clunker next to useless in comparison.
 
Last edited:
GK2001 said:
What kind of DSO was this? I haven't used a DSO in the last 25 years with a timebase that couldn't run "free". Just put the timebase in "Auto" mode and wind the trigger-level cursor off the screen. And I'd like to see an analog 'scope with a usable "roll" mode. Being a bit of an anlogue enthusiast, I have about 20 oscilloscopes and still resort to the analog occasionally, often because my 200 MHz / 2GSa/s Rigol DS1202CA can't match the bandwidth of my Tek 7904A.

The Rigol is over a decade old now and it has a pathetic memory depth which means it can't do things like properly display an AM or SSB modulation envelope, so nearly any of my analogue 'scopes with at least 20 MHz bandwidth are better for that task. However, all but the crappiest low-end DSOs these days have plenty of memory depth, allowing them to maintain high sampling rates at low timebase settings, so aliasing really isn't much of a thing with DSOs anymore. Oh, and the "digital phosphor" concept (trademarked by Tektronix and copied by nearly everyone else) of trace intensity modulation has got be be nigh on two decades old now. It was a bit crap back around about the year 2003 or 2004 when I first got the opportunity to asses a DSO with the capability, but things have come a long way since then, even for the budget machines. Skip to 6m 30s in the video here:

View attachment 1242581


But this is a vintage computer hardware forum, not a HAM radio forum. While nearly any old analog 'scope is better than no 'scope at all, when it comes to digital stuff, the single-shot timebase and storage capabilities of a DSO, even one as old as my Rigol DS1202CA, renders an old analog clunker next to useless in comparison.
Click to expand...
I was referring to the cheaper DSO's. A lot like the one suggested initially. I know the newer and better ones have improved no end , especially over the last 20 years. Still it is very difficult to beat the 400MHz Tek 2465B, unless you go to a very expensive DSO. Most 400MHz rated DSO's cannot display an 800 to 900MHz waveform like the 2465B can, they don't function well over their rated bandwidth. But like I say, it is like comparing apples & pears. And I did also say, that it is better to get something like a 100MHz rated new DSO than a beaten up analog oscilloscope. However, the better Analog scopes (like the 2465B & some others) still remain in a league of their own and the typical DSO cannot beat them, for certain applications.

Some new DSO's that can rival the 2465B have prices approaching $10k, similar to what the 2465B was back in the 1990's. I picked up my 2465b's for less than $800 each about a decade ago.

Generally though, the scopes I have been suggesting on this forum, for a start up scope, for those starting out, like the Hitachi V209 and V 509 20MHz & 50MHz are selling in the range of $100 to $200. It would be interesting to find any new DSO (or probaly used one) in that price range with that kind of performance that these analog scopes have.

It is also not a bad idea to master the Analog scope before moving to a DSO.

Another thing, but it might not be of too much interest, is that the Tek2465B is reagrded as "one of the last repairable Oscilloscopes" To the extent that it was supported by comprehensive manuals, schematics and spare parts. Unfortunately, a lot of modern test gear are "throw away items" not designed with repairs in mind as the manufacturers have moved to a built in obsolescence model for their business.
Here is a picture of a a 2465B showing a 0.98nS rising edge and some remarks by the seller about these scopes. To get anything that could perform this well in a new DSO, you had better be prepared to pay many thousands of $.

www.ebay.com

Tektronix 2465B Analog Oscilloscope for sale online | eBay

Find many great new & used options and get the best deals for Tektronix 2465B Analog Oscilloscope at the best online prices at eBay! Free shipping for many products!
www.ebay.com
 
Last edited:
Re: ds1054Z

To be fair this is quite long in the tooth now. It is the scope I wished I had bought... I actually bought a DS1052 before the 1054 came out and regret having only 2 channels. The critism of memory depth etc are also very valid but there is an assumption that for hobby use any budget will be very constrained.
 
I have a quite expensive logic probe but I never use it as the oscilloscope gives so much more information, just as easily and just as quickly.

£100 will give you all you will need for vintage PC's.

For me its a real no brainer.
 
Nivag Swerdna said:
I actually bought a DS1052........
Click to expand...

And for debugging vintage computer digital hardware, that, for the most part, is a much superior tool than any analog oscilloscope. If, right now, I was in the position of not owning an oscilloscope and looking for something to help fix old computers with on a limited budget, that is the kind of DSO that I'd be hunting for on the secondhand market.
 
Ok well this went off on a bit of a tangent haha, I am going to Likely order the Hantek unit below for now and if I get a real need for something better I will pickup one of its bigger brothers
https://www.amazon.ca/Hantek-Digita...e&qid=1655592558&sprefix=hantek,aps,99&sr=8-5

Back to the problem at hand what exactly am I looking for with regards to the video chips? Anything I should look for exactly or is it just test them one by one , leg at at time ?
 
GK2001 said:
And for debugging vintage computer digital hardware, that, for the most part, is a much superior tool than any analog oscilloscope. If, right now, I was in the position of not owning an oscilloscope and looking for something to help fix old computers with on a limited budget, that is the kind of DSO that I'd be hunting for on the secondhand market.
Click to expand...
A good challenge to verify the performance of a DS1052 and its suitability to logic circuit evaluation, would be to make a recording from a PET (if you have one).

The clock pulses that drive the 4 state machine JK flip flops, which generates the vertical drive pulses, are only about 100nS wide. They come in a group of 4, each about 1ms spaced and those groups of four are spaced by about 16.6mS, so it is an awkward configuration to scope them.

It would be interesting to see what the DS1052 would do with that. I found them difficult to resolve with a 50mHz scope , but the 2465B was able to do it to see the groups. When the timebase is set up to see one or the two groups of 4 they are so narrow as to be invisible on most analog scopes. It might be a case where the DSO could display them much better.

Generally with the analog scope, the way to view the pulses in an awkward pulse train like the one in the PET is to use the delayed independent timebase. You can create an intensified window and scroll it along over the narrow pulses, expand them out, count & examine them one by one and even scroll to the next group. Unfortunately not all analog scopes have this function (for example the V509 does but not the V502). So many people who use analog scopes are unfamiliar with the utility of this function for computer servicing, not realizing how useful it is. I would assume that the better DSO's would have an equivalent dual & delayed timebase function. They would have to have to be "as good". This is why I generally recommend the V509.

One trick a scope can be very useful for, is to measure the amplitude of a single narrow transient, which is too narrow to see on the scopes display. Good scopes have a calibrated trigger level and a trigger light, so this can be observed to check for a transient and the level adjusted until it just stops triggering. I used this method to find & measure very brief high voltage transients in automotive gear. In a sense it is like using the scope as a calibrated logic probe.
 
Last edited:
ryanhayward said:
Back to the problem at hand what exactly am I looking for with regards to the video chips? Anything I should look for exactly or is it just test them one by one , leg at at time ?
Click to expand...
That would be a good start. I try to work my way backwards to find specific faults, but you can also look for general issues such as voltages (and ripple) including voltage supply to the specific ICs, clock signals, CPU signals.

Hugo Holden said:
The clock pulses that drive the 4 state machine JK flip flops, which generates the vertical drive pulses, are only about 100nS wide. They come in a group of 4, each about 1ms spaced and those groups of four are spaced by about 16.6mS, so it is an awkward configuration to scope them.

It would be interesting to see what the DS1052 would do with that. I found them difficult to resolve with a 50mHz scope , but the 2465B was able to do it to see the groups. When the timebase is set up to see one or the two groups of 4 they are so narrow as to be invisible on most analog scopes. It might be a case where the DSO could display them much better.
Click to expand...
Now you got me curious to compare my DSO against the analogue one. Is the same circuit on a CBM 8032?
 
ryanhayward said:
Ok well this went off on a bit of a tangent haha, I am going to Likely order the Hantek unit below for now and if I get a real need for something better I will pickup one of its bigger brothers
https://www.amazon.ca/Hantek-Digital-Storage-Oscilloscope-Bandwidth/dp/B009H4AYII/ref=sr_1_5?crid=240D7RJQHI7EW&keywords=hantek+oscilloscope&qid=1655592558&sprefix=hantek,aps,99&sr=8-5

Back to the problem at hand what exactly am I looking for with regards to the video chips? Anything I should look for exactly or is it just test them one by one , leg at at time ?
Click to expand...

Just to jump in, that's what I bought last year when I was (with an awful lot of help) diagnosing and fixing my PET. Worked fine for me (although it was a lot cheaper when I bought it (UKP 33.49). There's some decent Linux software for it too - OpenHantek - if you want.

Colin.
 
You must log in or register to reply here.
Back
Top