Who sells tempcos?

[zipzap]

Hi again
Looking at al the shematics the people in the web provide, it seems like no (stable) vco can be build without a tempco resistor (is that true?).
Well, every company i have asked tells me they don´t know what a tempco is or that i have to by 100000+ of those.....
You guys seem to have build lots of vcos. where do you get them from? (btw, i´m in germany)
can´t have a decent modular without those 1v/oct vcos...

[Pehr]

http://www.precisionresistor.com/

[jksuperstar]

I don't know if this is your problem, but "Tempurate Compensating" resistor is what you are after.

[Pehr]

tempurate?

[Uncle Krunkus]

It all depends on how "stable" you want your VCO to be.
I prefer mine to have a sick sense of humour and be a few sandwiches short of a picnic!

[jksuperstar]

Oops tempurature

[Uncle Krunkus]

When the head chef tells you to dip 700 prawns in cold batter individually and fry them all until they are golden brown: -
That,..... is a tempura-ture!!

[zipzap]

yeah, how stable should it be? right now i use a 2 transistor, 2 cab 6 resistor pulse vco together with 4017 seq and a moog ladder. I love it.
But sometimes i wish i had something that i can tune to other musicians...
I found that german site http://de.farnell.com/jsp/endecaSearch/partDetail.jsp?SKU=732278&N=401
selling a 1k Temperature Sensing resistor for 26 cents.
Is that another name for the same thing? Or is it maybe the opposite?
It says something like +:3000ppm/°C. Well, the vco i want to build
http://www.musicfromouterspace.com/analogsynth/August2003VCO.html
wants 3350ppm whatever they are. Sounds like a problem?
cheers

[ian-s]

The paia 9700 VCO does not use a tempco.

http://www.paia.com/p9700s.htm

See site for circuit and description of how it works. It's a bit mystical for me.

[Scott Stites]

[zipzap]

Hi scott
so you think 3000ppm is to far away? Do you know how that would affect function of the whole thing? How come the 3000ppm resistor costs 26 cents and the one in the link 6,50? Maybe my link realy is somthing different.
And anyway, it would be some support for the people who deliver circuit and layout for free! Maybe i´ll order some when i need 3-4.
This paia thing does look interesting too. I´ll read the text tomorrow.

[Scott Stites]

I wouldn't say the 3000K would be too far out. Jorgen Bergfors (maker of the Bergfotron, one of the more colorful synths to be built) posted this a few years back on the SDIY list:

===
Hi DIYers.
I made some measurements this weekend on my ASM-1 clone VCO. I used two different tempco resistors and compared the compensation over an eight octave range. I also measured with an ordinary metal film resistor instead of the tempco. Measurements were done at 25 degrees and 38 degrees (centigrade). My CV quantizer was used to get repeatable control voltages.

Without compensation, the error is quite small att the high frequency range. At 7V CV (4186 Hz) it is 19 cents sharp. But with lower CVs the error increases linearly. At 0V CV (32,7 Hz) the error is a hefty 300 cent. No, that's not three dollars. It is three half note steps. Completely unacceptable, in other words.

With the 3000 ppm/C tempco, things looked completely different. Now it went slightly flat in the top range, when the temperature was rised. The error was -19 cent at 7V CV (4186 Hz). In this case the error didn't increase at lower frequencies. It got smaller. At the lowest frequency the error was so small that it couldn't be measured with mu frequency counter.

I also tried with a 3900 ppm/C tempco. This overcompensated slightly. At 7V CV (4186 Hz) the VCO went 30 cents flat. Lowering the frequency caused it to get flatter still. At the lowest frequency the error was -50 ppm.

For the tests I used the 1k SMD tempcos that Farnell sells. The expo converters were BC847BS dual transistors. They were potted together with the tempcos to keep them at the same temperature. Without potting even the smallest movement of air caused the frequency to wander.

The data (error in cents):

Frequency 0 ppm 3000 ppm 3900 ppm
32,7 305 0 -51
65,4 255 0 -39
130,8 218 -3 -39
261,6 180 -3 -45
523,3 135 -6 -39
1047 97 -10 -35
2093 57 -12 -32
4186 19 -17 -30

So as you see, the 3000 ppm tempco gives the optimum results. It is not totally perfect, but very close. Increasing the coefficient slightly (to the theoretically ideal 3300 ppm) would only mean that it would go slightly flat in the entire range.

/Jorgen
===

Make sure the tempcos you're looking at are postive coefficient tempcos (PTC), rather than negative coefficient (NTC).

Cheers,
Scott

[zipzap]

great! thats what i wanted to know. So i guess my temp sensing resistor +3000ppm is the right resistor. I hope its a ptc, i guess it should if it sais +3000

[Coriolis]

Sorry for digging this old thread up, but finding tempco's anywhere else but at Ray's keeps eluding me.

Doesn't help that I don't fully understand this coefficient thing (+3500 ppm, but what does it mean?).

I found these at Digikey:
http://www.nxp.com/pip/KTY81-2SERIES_4.html
They're 2K, and a bit over 1$ a piece, which I like better than 8$ for a PT146 from Ray. But are they any good? Looking at the datasheet, I don't see any mention of "ppm" anything... Are Precision Resistor stuff just better quality/precision, or are they simply something completely different from the KTY81's

Anybody know?

Ch

[toppobrillo]

the data sheet for the KTY parts shows they have a positive temp. coefficient, increasing [almost] linearly in resistance with positive change in temperature.

the t.c. is expressed in % Kelvin, which i ain't really familiar with..

3500ppm/deg. C means 'parts-per-million' in other words: 3500/1,000,000 or .0035 which i guess is the rate of exchange.

so, if your nominal R at room-temp is 2,000, would it follow that with a +3500ppm TC, when the ambient temperature rises 1 deg C, this R becomes 2,007 Ohms? then, if your ambient temp rises 1 degree more, we get 2,007x.0035, which equals 2014.0245... i could be way off here.

this is not a linear function. it's like interest accrument!! [arrgh]

well maybe i am way off. anybody know this stuff?

josh

[zipzap]

Sorry, i (still) don´t any better. I think you got it right. What i learned while searching for mine was that you get a lot of temperature sensors that are not well suited.
But if you do get a lable like 3500ppm that´s something. Also, and that´s what i do, if you get different values they can be compensated with normal resistors. Don´t know how exactly. Once had an exel sheet somewhere. I use a platinum PT1000 resistor. 3850ppm. In series 100R, 11k parallel. That is supposed to equal up to 3400ppm. Don´t know if it´s true.

[toppobrillo]

i think i have it wrong .. this comes from Rene Schmitz page about tempcos:

"To compensate the 1/T dependancy in the argument of the exp function one has to find a way to make the input voltage proportional to T, because T/T gives unity. The classical method is by a so called tempco resistor. That is a resistor made from a material whose resistance varies (almost) linear with absolute temperature. Platinum has about the right temperature coefficient, and so this material is often used. Now we see how it works in the figure. The voltage across the tempco-resistor is used as the input for the exponentiator."

and also this about the KTY devices:

"I came across the KTY 81-1xx series of devices made by Phillips. These are resistive temperature sensors, with a positive tempco. Only that the tempco behaves somewhat different than that of the Pt-resistors. When used with a series resistor of 1340 ohms, they act pretty much like a Pt-resistor of 2340 ohms.

The Figure 5 shows that the resistance of KTY+1.34K behaves closely to the Pt-resistor (straight line) within a temperature range of 0°C to 50°C. (The numbers for that diagramm have been pulled from the KTYs datasheet, and calculated for the PT-resistor.) The errors approach 0.2% at the borders. The temperature range is sufficient for normal applications."

so hey! maybe the KTY devices and your temperature sensing resistors will work fine..

here is the TEMP/RES. graphed for the PT146 PTC that ray likes and sells.

---

FIG4-1.gif
Description:
Filesize:	10.75 KB
Viewed:	8583 Time(s)

---

[Coriolis]

I must admit this is slightly over my head - would be easier if the numbers to look for were the same for every manufacturer. Prec Resistor uses degrees celcius/PPM (or vice versa?), and Phillips gives a resistance interval for a given temp. This confuses me, since I don't really understand too much of what is going on in the first place...

I have a vco with an expo pair made of glued-together to-92 transistors. When I put mu finger on them, pitch goes up. So obviously, the trick is to find a tempco, that counteracts that. But wouldn't different exp pairs have different characteristics? And so they would need different tempco's? Still, 3350 ppm tempco's are often touted as some sort of ideal value.

Gotta study this expo-pair thing a bit more.

Anyway, those KTY81/210's are cheap enough that I can get a few and try them out. If they don't work out, I'll buy from Ray (or directly from Precision Resistor).

C

[Uncle Krunkus]

Every unique set of components will have slightly different characteristics, and that's why you'll find a trimpot in there somewhere that will need to be fine tuned. I think when it comes to compensating with tempcos, you're dealing with the temperature change of the junction in the transistor, which is fairly predictable, but not exact.

[Coriolis]

[Coriolis]

Btw: Has anyone here tried buying directly from Precision Resistor?

C

[toppobrillo]

hey

yeah i inquired to them about a year or so ago, they have a minimum to buy.. they have this in stock:

PT146 1000ohm 1.000% TC3500+ppm $3.10ea

you gotta order like 8 orso at the minimum.