4
\$\begingroup\$

The CanoScan 8800F uses a proprietary 32VDC AC adapter that's no longer manufactured. A Flickr user did the trial and error to find out how to modify another 32VDC AC adapter for use with the 8800f. It seems simple enough. One pin carries the 32VDC of current, and other pin carrying current starts off carrying current through a resistor only to have that resistor shorted after a few seconds so that the full 32VDC of current can be delivered. Could someone advise me what the parts and schematics would be to build "a time delay relay assembly" to "short a 1.5k ohm resistor" "after a few seconds" of the scanner being powered on?

I tried messaging the user to get a more detailed answer, but to no avail. I figure he assumed that if you have more electronics knowledge his solution would be easy to interpret. Here is his original post from Flickr:

"Here is the real deal. I purchased Canon Canoscan 8800F with no power adapter. After extensive testing here is how it works. Looking at the back of the scanner the left pin will be referred to as pin 1, center pin as pin 2 and right pin as pin 3. Now, I had to build a circuit which adds a time delay. Let me tell you how to get your scanner to work. Pin 2 is ground, Pin 3 is 32VDC. Pin 1 wants the same as pin 3 (32VDC) with the following exceptions: when pin 3 receives 32VDC pin 1 will receive the same 32VDC BUT in series must be 1.5k ohms. Then, after a few seconds, the 1.5k ohm resistor must be shorted allowing the full 32VDC to be applied to pin 1. I built a time delay relay assembly to facilitate this function and now the scanner will power on when the power button is pressed. Hope this helps."

https://www.flickr.com/groups/767295@N20/discuss/72157627293403452/72157647465205394

Thanks.

\$\endgroup\$
  • \$\begingroup\$ Does anyone know the purpose of such functionality? Some kind of in-rush current limiting? \$\endgroup\$ – dext0rb Mar 31 '15 at 2:50
  • \$\begingroup\$ Sounds like it should be doable with a relay, a 555 timer IC, a 5V regulator (to supply the 555 IC) and associated resistors/capacitors. \$\endgroup\$ – Li-aung Yip Mar 31 '15 at 3:12
  • 1
    \$\begingroup\$ So the scanner does startup checks, loads the power supply, which then ramps voltage on pin one? It could be inrush limiting, but this sounds more procedural than that. Like signaling power_good or something. \$\endgroup\$ – Sean Boddy Mar 31 '15 at 6:17
  • \$\begingroup\$ Or you know, just a bs reason to limit the second hand market and price gorge on replacement parts. \$\endgroup\$ – Passerby Mar 31 '15 at 14:04
  • \$\begingroup\$ @Passerby LOL, I was hoping it's not actually that. But yeah, probably. \$\endgroup\$ – dext0rb Mar 31 '15 at 16:22
3
\$\begingroup\$

Sounds like a job for "time delay relay"/"delay-on relay". You have three options:

|improve this answer|||||
\$\endgroup\$
3
\$\begingroup\$

If you want a time delay circuit to build yourself, here's one for you -- the TL431 is a wonderfully underrated part, capable of serving as a comparator/reference combo in addition to being a programmable shunt reference. The IRF7540 can be replaced by any standard gate (i.e. Rds(on) at 10V Vgs) power PMOSFET, and the timing capacitor should be a stable, low-leakage type such as a polyester film or a good grade of tantalum cap -- the current values are chosen for a 3s delay-ON in the simulation and will need to be tweaked for a real TL431. Also, the inrush limiter resistor should be rated for at least 1W.

schematic

simulate this circuit – Schematic created using CircuitLab

As to how this thing works -- U1 (represented by the cobbled-together macromodel in the dotted box because CircuitLab, in their infinite wisdom, does not provide a TL431) is used open-loop as a comparator, turning ON when Ctiming charges to 2.5V via Rtiming, while R2 holds Q1 and Q2 off until U1 turns on. (SW-IC is used to set the initial condition for Ctiming to 0V as CircuitLab lacks SPICE's .IC command.) When U1 turns on, it pulls its cathode to about 2V above the anode -- D1 adds another 18V of voltage drop atop that to keep Q1s gate from being pulled beyond its absolute voltage rating (i.e. to keep Q1 from going bang). Once U1 is solidly ON, Q1 and Q2 both switch ON, shorting out Rlimiter (i.e. the 1.5kΩ resistor from the spec) and pulling U1's adjust terminal HIGH through R1 in order to prevent noise from causing the output to chatter (in other words, Q2 and R1 turn the TL431 into a crude Schmitt trigger). RL serves as a demonstration load.

|improve this answer|||||
\$\endgroup\$
2
\$\begingroup\$

@ThreePhaseEel has a good approach; I'd simplify the circuitry somewhat, because there doesn't seem to be need for accuracy or sharp turnon.

schematic

simulate this circuit – Schematic created using CircuitLab

If you want a sharper turnon, R4 can be replaced with a 24V zener diode.

|improve this answer|||||
\$\endgroup\$

Not the answer you're looking for? Browse other questions tagged or ask your own question.