Re: Anyone build a ZVC (Zeprom Voltage Converter) Continued - 2

[Dan Grelinger]

I have continued to flush out the parts manifest, but am not quite done. There are two parts I need help with; one needs a little help with specs, and the other a source (where I can buy less than thousands).

But first, here are the specs I have for the ZVC and ZEPROM. This might be helpful in specing out the coil.

When the ZVC is turned on (generating 12.5V), but the ZEPROM is not programming itself, the measured current from the HP-41 battery supply is 4.9 mA.

When the ZEPROM starts programming itself, the total measured combined current for both ZEPROM and ZVC is 14.6 mA average, 15.1 mA peak. (HP-41 power supply is 6 volts, the ZVC provides a 12.5 volt supply to the ZEPROM for programming)

I think I have a set of good specs for the parts, except for one, the coil. What I know is that it is a 220 uH Ferrite Core Inductor (Thank you, Deigo!) and that I measure a resistance across it of approximately 2.2 Ohms (after subtracting lead and contact resistance). When I searched digikey's vast catalog of ferrite core through-hole inductors, I find several that have this approximate DC resistance, and with what appears to me to be a reasonable peak current specification (160 mA - 230 mA). Unfortunatly, none of them are stocked, and minimum quantities are 500, for a total price of $56.88. What is a minimum max. current spec that I should use? Anything over 100 mA would seem to be reasonable, given the ZVC measured current values.

The 1-ohm, 1% tolerance, 100 ppm/C spec'ed resistors are hard to find in reasonable quantities as well. Digikey has them, but I have to buy 2500 or 5000 minimum quantity. I don't need more than 30 of them.

Update Part Manifest (digikey part numbers):

(1) IC1- MC34063 P1 Voltage Booster (296-17766-5-ND)

(3) D1,2,3 - 1N4001 Diode (1N4001-TPMSCT-ND)

(1) Cout - 470 uF Capacitor (565-1660-ND)

(1) Cin - 100 uF Capacitor (P963-ND)

(1) CT - 1000 pF Capacitor? (BC1072CT-ND)

(1) C1 - 0.1 uF Capacitor (BC1148CT-ND)

(3) R3,4,5 - 1 Ohm Resistors, 1% Tolerance 100 ppm/C (Need source)

(1) R6 - 150 Ohm Resistor, 1% Tolerance (P150CACT-ND)

(1) R1 - 2.4 kOhm Resistor, 1% Tolerance (P2.40KCACT-ND)

(1) R2 - 22 KOhm Resistor, 1% Tolerance (P22.0KCACT-ND)

(1) 220 uH Ferrite Core Indcutor (Need Source)

(1) Switch

(1) Double Sided PCB

(1) HP-41 Module contacts and housing (anyone know a source?)

Thanks for any and all help!

Dan

[Dave Shaffer (Arizona)]

Maybe the switching power supply gurus can comment, but I wonder if you really need the 1% tolerance on the resistors.

The individual 1 or 2 digit significant figure values are all standard for 5% (and most for 10%) tolerance, and the 100PPM for the 1 ohm seems way more than necessary: if it is within 1% to start with, it would take a 100 degree C temperature change to change the resistance by another 1% ! Maybe the original builder just had a bunch of fancy resistors lying around and used them because they were convenient.

The capacitors are also all standard values, and are probably good to 10% if you are REAL lucky. Same with the inductor - another standard 10% tolerance value, and again, I'll bet it's not necessarily all that close to 220uH. Maybe you could even wind your own, if you can find a compatible core and fine enough wire.

This still might not help with the 1 ohm resistors though - that's a pretty low resistance value.

[Richard Ottosen]

Quote:

---

Maybe the switching power supply gurus can comment, but I wonder if you really need the 1% tolerance on the resistors.

---

The only resistors that truly need to be 1% tolerance are the ones that set the output voltage -- the 2.4K and the 24K.

If you still want to do 1% resistors for the 1 ohm parts then you can use surface mount resistors:

http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=311-1.00FRCT-ND

It is not as hard as it sounds. For prototyping, you use a board with pads on 0.1" centers. Solder the first of the three resistors to two adjacent pads. Then simply solder the other two resistors on top of the first one. Now you can connect wires to the ends of the stacked resistors.

However, if I were doing it, I would use 5% resistors since this value does not seem critical. Any one in this list would work:

http://search.digikey.com/scripts/DkSearch/dksus.dll

Quote:

---

Maybe you could even wind your own, if you can find a compatible core and fine enough wire.

---

Making your own coil is a _lot_ harder than it sounds. The makings for a proper coil are harder to get than the coil. Designing and winding an accurate coil is even harder. :-(

This should work for your inductor even though it is a bit large:

http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=M9997-ND

-- Richard

[Garth Wilson]

Zengrange might have had some policy to just use 1% on everything just to make it simple since the resistors didn't account for much of the overall cost. Most 5% resistors are within a percent anyway, and having the current-sensing accuracy is just not that critical. I'd put in 5% and be done with it.

For the inductor, the peak current can be several times as high as the regulator's output current, but how much will depend on several factors which are probably outlined and graphed in the data sheet. It sounds like your 100mA inductor can handle a lot more than enough. You can get suitable ones in singles any day of the week from lots of stocking distributors.

Edited: 27 July 2008, 12:46 a.m.

[Scott Newell]

Quote:

---

The 1-ohm, 1% tolerance, 100 ppm/C spec'ed resistors are hard to find in reasonable quantities as well. Digikey has them, but I have to buy 2500 or 5000 minimum quantity. I don't need more than 30 of them.

---

Nah, those are easy. Try Mouser. 660-MF1/4DC1R00F shows as in stock at $0.04/each.