[cctalk] Re: Hot Video Shift Register on VT100

-----Original Message----- From: Charles Morris via cctalk <cctalk(a)classiccmp.org> Sent: 30 November 2025 15:24 To: cctalk(a)classiccmp.org Cc: Charles Morris <charlesmorris800(a)protonmail.com> Subject: [cctalk] Re: Hot Video Shift Register on VT100 Keep in mind that ringing (artifact caused by the inductance of the scope probe ground lead) can result in a false display of overshoots... How short is your ground lead?

-Charles On 11/30/25 08:30, Rob Jarratt via cctalk wrote: > I have traced the clock signal back. The MAS CLK input on the DC011 > also has the negative spikes. This is coming from a 74S157. The input > to that is an oscillator and that too has the negative spikes. > > So, it seems the 74S157 is just passing through the inputs. > > There is a capacitor across pins 4 and 2 of the oscillator (Vcc and > Gnd), would a failure there cause this kind of negative spike? It > seems unlikely to me, but would be happy to be corrected. > > The oscillator is rated at 24.0734MHz. I can find lots rated at > exactly 24MHz and I guess the difference in frequency is small enough > not to matter, right? > > I will order the part and also a new capacitor for it in case that is > where the fault lies and see. > > Thanks > > Rob > > >> -----Original Message----- >> From: Rob Jarratt via cctalk &lt;cctalk(a)classiccmp.org&gt; >> Sent: 30 November 2025 12:57 >> To: 'General Discussion: On-Topic and Off-Topic Posts' >> &lt;cctalk(a)classiccmp.org&gt; >> Cc: 'Hans-Ulrich Hölscher' &lt;vaxorcist(a)googlemail.com&gt;om>; Rob Jarratt >> &lt;robert.jarratt(a)ntlworld.com&gt; >> Subject: [cctalk] Re: Hot Video Shift Register on VT100 >> >>> -----Original Message----- >>> From: Hans-Ulrich Hölscher via cctalk &lt;cctalk(a)classiccmp.org&gt; >>> Sent: 29 November 2025 09:17 >>> To: General Discussion: On-Topic and Off-Topic Posts >> &lt;cctalk(a)classiccmp.org&gt; >>> Cc: Hans-Ulrich Hölscher &lt;vaxorcist(a)googlemail.com&gt; >>> Subject: [cctalk] Re: Hot Video Shift Register on VT100 >>> >>> Hi Rob, >>> this is not intended to indicate the cause or solution, but merely >>> to help analyze how critical your IC temperature is: Did you >>> actually measure the temperature of the IC? I recently bought an >>> inexpensive infrared thermometer >>> (like:https://www.ebay.co.uk/itm/156810933790) >>> that can measure the temperature of small areas relatively accurately. >>> I made the experience that I cannot judge temperatures very well >>> with my >> fingers. >>> According to Wikipedia, TTL ICs are specified for temperatures up to >>> 70 °C. This >>> temperature should obviously be avoided, but my fingers already >>> perceive anything above about 50 °C as "hot". >>> >>> I hope this can help a litte! >> Yes, I have an IR Thermometer and the chip gets to about 40 celsius. >> I > realise >> that this is well below the maximum rating, but I can smell the heat >> and > it >> worries me a bit. If I knew this was expected then I would be OK with >> it, > but it >> feels anomalous and I have felt the chip to be hotter than this sometimes. >> The observation about the negative voltages also concerns me a bit, > although >> I do have spares for the 74S299 should I need them. >> >> It looks now like the DC011 is just producing a DOT CLK output with > negative >> spikes. Maybe this is why later revisions introduced an inductor or >> maybe > it is >> just a bad DC011 or something is wrong with what is around the DC011. >> >> I am going to look at the DC011 inputs to see if anything looks >> amiss. The other thing I would like to do is to see what happens if I >> can add an > inductor >> to the output of the DC011. But I don't know how to read the spec on >> the printset to know what I might need. The printset says "BEAD, >> FERRITE .1380DX.380LG", can anyone tell me what this might mean in >> terms of an inductor that I could try? >> >> Thanks >> >> Rob >> >> >>> Ulli >>> >>> Am Sa., 29. Nov. 2025 um 09:44 Uhr schrieb Henk Gooijen via cctalk < >>> cctalk(a)classiccmp.org&gt;gt;: >>> >>>> Rob, >>>> >>>> I have a few silly thoughts, but maybe they can be useful. >>>> The negative going signal is something I would not be happy with, >>>> as it may damage the IC (over time). >>>> TTL does not like negative voltages on their pins. >>>> Is there between the DC011 (signal generating) and the LS299 >>>> (signal >>>> "user") wiring, or are they on the same PCB? >>>> In case of wiring, capacitive coupling may cause negative going >> voltages. >>>> Could a clamping diode with an as low as possible Vf (< 0.3 V) help >>>> to reduce the negative voltage? It is not a permanent solution, >>>> because the >>>> DC011 might be a bit unhappy with that diode, but for a test it >>>> might help to draw a conclusion. >>>> >>>> If all does not help, in getting the LS299 running cooler, consider >>>> some temperature conductive paste and a small "IC heatsink". Tie >>>> the heatsink onto the IC with a piece of waxed lacing. The IC can >>>> radiate more efficiently the generated heat, so that it can run >>>> cooler. This could drop temperature with some 10 degrees. >>>> >>>> Success hunting down this issue, >>>> Henk >>>> >>>> -----Oorspronkelijk bericht----- >>>> Van: Rob Jarratt via cctalk &lt;cctalk(a)classiccmp.org&gt; >>>> Verzonden: zaterdag 29 november 2025 08:11 >>>> Aan: 'Doug Jackson' &lt;doug(a)doughq.com&gt;om>; rob(a)jarratt.me.uk >>>> CC: 'General Discussion: On-Topic and Off-Topic Posts' < >>>> cctalk(a)classiccmp.org&gt;gt;; Rob Jarratt &lt;robert.jarratt(a)ntlworld.com&gt; >>>> Onderwerp: [cctalk] Re: Hot Video Shift Register on VT100 >>>> >>>> Yes, it is a negative voltage. I measured this by attaching the >>>> ground lead to the ground connection on the 74S299. >>>> >>>> >>>> >>>> The DC011 that is producing the CLK signal is socketed, but the one >>>> on my >>>> VT102 isn’t. I would try swapping them to see what happens, but I >>>> don’t really want to touch a fully working board. >>>> >>>> >>>> >>>> Thanks >>>> >>>> >>>> >>>> Rob >>>> >>>> >>>> >>>> From: Doug Jackson &lt;doug(a)doughq.com&gt; >>>> Sent: 28 November 2025 22:59 >>>> To: rob(a)jarratt.me.uk >>>> Cc: General Discussion: On-Topic and Off-Topic Posts < >>>> cctalk(a)classiccmp.org&gt; >>>> Subject: Re: [cctalk] Re: Hot Video Shift Register on VT100 >>>> >>>> >>>> >>>> Is that a negative excursion at the bottom of the waveform? now >>>> low >>>> - looks like -0.3v to me - With '"logic, that will likely turn on >>>> the protection diodes on the inputs, making them warm. >>>> >>>> >>>> >>>> Double check where the ground actually is on the CRO display. >>>> >>>> >>>> >>>> It's also interesting that the input doesn't even reach 4v, let >>>> alone >> 5v. >>>> Whats driving that? >>>> >>>> >>>> >>>> >>>> >>>> Kindest regards, >>>> >>>> >>>> >>>> Doug Jackson >>>> >>>> >>>> >>>> em: doug(a)doughq.com <mailto:doug@doughq.com> >>>> >>>> ph: 0414 986878 >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> On Sat, 29 Nov 2025 at 08:52, Rob Jarratt >>>> &lt;robert.jarratt(a)ntlworld.com <mailto:robert.jarratt@ntlworld.com> > >> wrote: >>>> I checked the CLK input for the 74S299 with the 74S299 removed and >>>> the signal is still spiky and looks like this: >>>> https://rjarratt.wordpress.com/wp-content/uploads/2025/11/vt102-74s >>>> 2 >>>> 99 >>>> -clk-signal.png >>>> >>>> >>>> >>>> I removed the other load on this signal, the DC012, and the CLK >>>> signal still looked spiky. I also put back the 74S299 (now >>>> socketed) with >>>> DC012 removed and it still gets hot. I tried another new 74S299 >>>> just in case I had a bad one and that also got hot without a load >>>> on the >> output. >>>> >>>> >>>> I am thinking that it is either the spiky input or it is just >>>> expected to get this hot. >>>> >>>> >>>> >>>> Regards >>>> >>>> >>>> >>>> Rob >>>> >>>> >>>> >>>> From: Doug Jackson &lt;doug(a)doughq.com <mailto:doug@doughq.com> > >>>> Sent: 25 November 2025 01:15 >>>> To: rob(a)jarratt.me.uk <mailto:rob@jarratt.me.uk> >>>> Cc: General Discussion: On-Topic and Off-Topic Posts < >>>> cctalk(a)classiccmp.org <mailto:cctalk@classiccmp.org> > >>>> Subject: Re: [cctalk] Re: Hot Video Shift Register on VT100 >>>> >>>> >>>> >>>> The oscilloscope will show you the actual driving waveform. >>>> >>>> >>>> >>>> You can then see the level of the spikes >>>> >>>> >>>> >>>> Kindest regards, >>>> >>>> >>>> >>>> Doug Jackson >>>> >>>> >>>> >>>> em: doug(a)doughq.com <mailto:doug@doughq.com> >>>> >>>> ph: 0414 986878 >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> On Tue, 25 Nov 2025 at 10:48, Rob Jarratt >>>> &lt;robert.jarratt(a)ntlworld.com <mailto:robert.jarratt@ntlworld.com> > >> wrote: >>>> >>>> >>>>> -----Original Message----- >>>>> From: Doug Jackson via cctalk &lt;cctalk(a)classiccmp.org <mailto: >>>> cctalk(a)classiccmp.org&gt; > >>>>> Sent: 24 November 2025 23:04 >>>>> To: General Discussion: On-Topic and Off-Topic Posts < >>>> cctalk(a)classiccmp.org <mailto:cctalk@classiccmp.org> > >>>>> Cc: Doug Jackson &lt;doug(a)doughq.com <mailto:doug@doughq.com> > >>>>> Subject: [cctalk] Re: Hot Video Shift Register on VT100 >>>>> >>>>> Another test would be to disconnect the output and see if that >>>>> runs the device cooler. Maybe the load, while not a short, is >>>>> simply too great >>>> for the >>>>> device. >>>> I think I can try this, the DC012 it drives is socketed. >>>> >>>>> But I am leaning towards the inputs containing spikes that need to >>>>> be dissipated as heat in clamping diodes. A decent CRO on the >>>>> device input >>>> pins >>>>> while it is out of circuit will show that as well. >>>> I guess CRO means Cathode Ray Oscilloscope? Not sure what you are >>>> suggesting I try here? >>>> >>>> >>>>> Kindest regards, >>>>> >>>>> Doug Jackson >>>>> >>>>> em: doug(a)doughq.com <mailto:doug@doughq.com> >>>>> ph: 0414 986878 >>>>> >>>>> >>>>> >>>>> >>>>> On Tue, 25 Nov 2025 at 09:52, Peter Coghlan via cctalk < >>>>> cctalk(a)classiccmp.org <mailto:cctalk@classiccmp.org> > wrote: >>>>> >>>>>>> Sorry if I wasn't clear in my earlier reply, but there were some >>>>>>> answers inline. The summary is that Vcc looks fine, there does >>>>>>> not seem to be any short on the outputs of the shift register >>>>>>> and the chip seems to be >>>>>> working >>>>>>> correctly. It just gets hot and I wonder if this could be due to >>>>>>> the CLK input being a bit spikier on the VT100 than on my > VT102? >>>>>>> Could this be >>>>>> why >>>>>>> later revisions of the VT100 introduced an inductor on the DOT >>>>>>> CLK output from the DC011? >>>>>>> >>>>>> Perhaps one way to test the hypothesis would be to insert an >>>>>> inductor like those used on the later revisions in the DOT CLK >>>>>> output and check whether >>>>>> a) everything still works and b) the shift register runs cooler? >>>>>> >>>>>> Regards, >>>>>> Peter Coghlan. >>>>>> >>>>