My RESET-7 POR upgrade for the Simmons SDS7 was designed to extend effective battery life when the SDS7 has been converted to lithium battery backed up memory. This might not seem like a big deal but please allow me to explain...
The Simmons SDS7 uses a CD4011 to buffer, invert and interface a signal derived from the power supply on power-up and power-down and which provides the chip enable (CE) signal for the SRAM and RST (reset) for the processor. It's an important part of the Power-On-Reset (POR) circuitry in the SDS7. Since the device is on all the time, it's powered by the memory back-up battery when the SDS7 is switched off. Yes, that's right! The 'memory back-up battery' doesn't just back up the memory! CE on the 6116s needs to be high, when SDS7 is switched off. To do that, the CD4011 needs to be on.
The SRAM in the SDS7 comprises ten 6116 chips. These have a minimum memory retention voltage of 2V so theoretically, the memory back-up voltage can run down to this value and memory will be retained.
Oh, if only life would be that simple...
The minimum operating voltage of the CD4011 however, is 3V. If the memory back-up battery voltage falls below 3V, memory integrity will be compromised and corruption may occur.
Bypassing the SDS7's original POR circuitry, my Dark Matter replacement power supply for the Simmons SDS7 gets around this problem with a completely new POR design. Apart from being based around a power management IC, Dark Matter's POR circuit uses a SN74HC14N hex inverter to buffer, invert and distribute the RESET pulse and to maintain CE as required, for the 6116s. The big deal is that the SN74HC14N has a minimum operating voltage of (you guessed it), 2V.
Having a lower current consumption than the CD40111 and Schmitt-trigger inputs, also constituted towards my decision to use the SN74HC14N. For a simple hex inverter, it's a the perfect device in this situation.
"Hold on a minute, Alex. The CD4011 is a quad 2-input NAND gate and the SN74HC14N is a hex inverter. How can you replace one with the other?" I hear you ask.
Okay, let's talk Simmons SDS7 RESET for a minute...
The POR signal in the SDS7 is derived from a simple R/C network taken off the power supply, post rectifier / pre-regulator. The output of the R/C network is fed into a CD4011 quad 2-input NAND gate where it's conditioned, buffered, inverted and distributed to the CPU and memory boards. RESET (RST) to the CPU and chip enable (CE) to the SRAM chips, to be precise.
Only three of the four NAND gates of the CD4011 are used and only two of them are used for RST and CE. The two inputs of each NAND gate are joined together which means that only the 'N' of the NAND is used and not the 'AND', if you know what I mean. We can therefore substitute an inverter in-place of each NAND gate.
Of course my revised RESET circuit only uses three of the six inverters in the SN74HC14N but it's still a very perfect alternative. Hey, sometimes you have to look at what's actually going on, right?
Using just a power management chip might have given SDS7 a heart-attack. Putting a logic gate on the backend however, fools SDS7 into thinking that everything is as it was in 1983 (perhaps). 🤣
The installation of Dark Matter is however, very involved and so it's not going to be for everyone. To allow for simple conversion of SDS7's memory back-up to lithium, I therefore designed Pleiades. Unfortunately, Pleiades doesn't address the problem described here and so I decided to come up with a solution.
Since the CD4011 is nearly always socketed (in fact, I've never seen a SDS7 in which the CD4011 isn't socketed), my idea was to make a simple adapter which would be easy to fit once the CD4011 had been removed from its socket. There was one major problem, though. The CD4011 (otherwise referenced as IC1) couldn't be in a more inaccessible place. I wanted this to be easy but being right at the back of the SDS7 chassis, kind of underneath the vertically mounted rear socket PCB and surrounded by other components, meant that it would be anything but!
There wasn't enough room to drop in a conventional, long pin vertical adapter so my first version of RESET-7 had right-angled pins on the back, the idea being that the adapter would be mounted vertically. Unfortunately, room was so tight that even this didn't fit.
So, my next option was to mount RESET-7 on the rear PCB and connect it to IC1 socket via a short IDC cable. Should I need to use this mounting option, I'd made a couple of 3.2mm holes on the RESET-7 PCB to accommodate a two self-adhesive PCB stand-offs.
My RESET-7 POR upgrade for the Simmons SDS7 was designed with maintenance in mind so apart from having the same orientation as the original device, the SN74HC14N is also socketed, easy to access and the PCB has a couple of conveniently located test points, thus keeping an eye on voltage that's hitting the SN74HC14N when SDS7 is powered on or off, is simple.
You may be wondering about the relevance of the Schmitt trigger aspect of the SN74HC14N's inputs that I mentioned earlier. Well, unlike the old CD4011, the outputs of the SN74HC14N will only switch state when the respective inputs exceed a specific threshold. This means that no POR signal will be generated from spurious activity and so the possibility of a 'false' RESET will be greatly minimised.
So, if you're considering a lithium back-up battery conversion on your SDS7, I would firstly recommend Pleiades. If on the other hand, you're knowledgeable and confident to try something yourself, then do yourself a favour and please do consider RESET-7. Offering reduced current consumption from the battery when SDS7 is switched off and having a minimum operating voltage of 2V, your lithium battery (or batteries) will simply last longer.RESET-7 POR upgrade for the Simmons SDS7 is now offered as a highly recommended option when you buy Pleiades.
UPDATE - 18th June 2025
I've received an enquiry asking if RESET-7 would work in a SDS7 which has NOT been converted to lithium battery back-up. I haven't actually tested this but I honestly don't see why not.
The theory of operation remains the same but since a Ni-Cd needs to be regularly charged anyway, I'm not sure what advantage would be gained, other than having a very cool POR backend.