CEM3340 Pin 3 Considerations

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: luketeaford, Joe., lisa, Kent

Post Reply
Studioalethea
Wiggling with Experience
Posts: 265
Joined: Mon Dec 23, 2019 2:20 am
Location: Seattle

CEM3340 Pin 3 Considerations

Post by Studioalethea » Wed Jul 29, 2020 2:24 pm

Hey All!

I think I may be making a simple Ohm's law miscalculation OR have a misunderstanding about the interaction of the chip on voltage...

I am working on the VCO Maximus circuit from Thomas Henry, originally for +/-15V and working it for Eurorack levels:
https://www.birthofasynth.com/Thomas_He ... %203.5.pdf

It is my understanding that the goal is to have -5V at Pin 3 of the CEM3340. Sometimes designers use a voltage regulator of some kind, but I didn't really want to go that direction.

The original datasheet accomplishes this with an 820R (R4) resistor to -15V.

I therefore assumed that I would have to use a 620R resistor to -12V power.

However my voltage measures around -7V in practice.

I checked my design against others that exist and LMNC does use a 620R resistor:
https://www.lookmumnocomputer.com/cem-3340-diy-simple

However, Fonitronik's design is a little confusing: His BOM indicates a 1.2K resistor for R4, while his schematic shows the original 820R.
http://fonitronik.de/application/files/ ... AXIMUS.pdf

So, what's the deal? What's the part of Ohm's law I am misunderstanding OR what is going on in the chip to influence this calculation?

Thanks all!

socom93
Common Wiggler
Posts: 63
Joined: Sun Mar 24, 2019 1:19 pm
Location: France

Re: CEM3340 Pin 3 Considerations

Post by socom93 » Wed Jul 29, 2020 2:39 pm

The datasheet says there is an internal6.5v zener diode. So no miscalculation they also give how to calculate the value of the limiting resistor.

User avatar
infinitemachinery
Common Wiggler
Posts: 144
Joined: Sat Jan 05, 2019 11:01 am
Location: Ashland, Oregon
Contact:

Re: CEM3340 Pin 3 Considerations

Post by infinitemachinery » Wed Jul 29, 2020 2:51 pm

Having a regulated -5V supply is best. Pretty simple circuit using an L79L05

Studioalethea
Wiggling with Experience
Posts: 265
Joined: Mon Dec 23, 2019 2:20 am
Location: Seattle

Re: CEM3340 Pin 3 Considerations

Post by Studioalethea » Wed Jul 29, 2020 4:11 pm

infinitemachinery wrote:
Wed Jul 29, 2020 2:51 pm
Having a regulated -5V supply is best. Pretty simple circuit using an L79L05
Presumably all this adds is the regulator and a few wires, right? Feed in -12V and out pops a pretty steady -5V?

Borogove
Learning to Wiggle
Posts: 37
Joined: Sun Feb 25, 2018 10:56 pm

Re: CEM3340 Pin 3 Considerations

Post by Borogove » Wed Jul 29, 2020 4:28 pm

socom93 wrote:
Wed Jul 29, 2020 2:39 pm
The datasheet says there is an internal6.5v zener diode. So no miscalculation they also give how to calculate the value of the limiting resistor.
Yep, Ree is a current limiter into the zener, not a resistive divider.

The AS3340 datasheet recommends (Vee-7.4) / 0.008, the CEM3340 datasheet recommends (Vee-7.2) / 0.008. For -12V, that's 575 ohms for the AS, 600 ohms for the CEM, so you should be good with your 620 ohms. The 3340 will run cooler if you give it a -5V supply instead of the -12, though.

Borogove
Learning to Wiggle
Posts: 37
Joined: Sun Feb 25, 2018 10:56 pm

Re: CEM3340 Pin 3 Considerations

Post by Borogove » Wed Jul 29, 2020 4:30 pm

Studioalethea wrote:
Wed Jul 29, 2020 4:11 pm
infinitemachinery wrote:
Wed Jul 29, 2020 2:51 pm
Having a regulated -5V supply is best. Pretty simple circuit using an L79L05
Presumably all this adds is the regulator and a few wires, right? Feed in -12V and out pops a pretty steady -5V?
Usually a couple of caps as well to keep things smooth, but yeah.

Borogove
Learning to Wiggle
Posts: 37
Joined: Sun Feb 25, 2018 10:56 pm

Re: CEM3340 Pin 3 Considerations

Post by Borogove » Wed Jul 29, 2020 4:31 pm

Borogove wrote:
Wed Jul 29, 2020 4:28 pm
socom93 wrote:
Wed Jul 29, 2020 2:39 pm
The datasheet says there is an internal6.5v zener diode. So no miscalculation they also give how to calculate the value of the limiting resistor.
Yep, Ree is a current limiter into the zener, not part of a resistive divider.

The AS3340 datasheet recommends (Vee-7.4) / 0.008, the CEM3340 datasheet recommends (Vee-7.2) / 0.008. For -12V, that's 575 ohms for the AS, 600 ohms for the CEM, so you should be good with your 620 ohms. The 3340 will run cooler if you give it a -5V supply instead of the -12, though.

Studioalethea
Wiggling with Experience
Posts: 265
Joined: Mon Dec 23, 2019 2:20 am
Location: Seattle

Re: CEM3340 Pin 3 Considerations

Post by Studioalethea » Wed Jul 29, 2020 4:39 pm

Borogove wrote:
Wed Jul 29, 2020 4:28 pm
socom93 wrote:
Wed Jul 29, 2020 2:39 pm
The datasheet says there is an internal6.5v zener diode. So no miscalculation they also give how to calculate the value of the limiting resistor.
Yep, Ree is a current limiter into the zener, not a resistive divider.

The AS3340 datasheet recommends (Vee-7.4) / 0.008, the CEM3340 datasheet recommends (Vee-7.2) / 0.008. For -12V, that's 575 ohms for the AS, 600 ohms for the CEM, so you should be good with your 620 ohms. The 3340 will run cooler if you give it a -5V supply instead of the -12, though.
In here somewhere is clearly my confusion— the current limiting resistor vs the voltage divider. Where is the voltage divider created in the datasheet schematic? Between the 600/620R resistor and... the trimmer?

FrankLacy
Learning to Wiggle
Posts: 2
Joined: Wed Mar 04, 2020 1:41 pm

Re: CEM3340 Pin 3 Considerations

Post by FrankLacy » Wed Jul 29, 2020 5:33 pm

An old scan of a CEM3340 datasheet states that the internal Zener is 6.5V +-10%; a measurement of -7V falls in that range.

Borogove
Learning to Wiggle
Posts: 37
Joined: Sun Feb 25, 2018 10:56 pm

Re: CEM3340 Pin 3 Considerations

Post by Borogove » Wed Jul 29, 2020 5:48 pm

Studioalethea wrote:
Wed Jul 29, 2020 4:39 pm
Borogove wrote:
Wed Jul 29, 2020 4:28 pm
socom93 wrote:
Wed Jul 29, 2020 2:39 pm
The datasheet says there is an internal6.5v zener diode. So no miscalculation they also give how to calculate the value of the limiting resistor.
Yep, Ree is a current limiter into the zener, not a resistive divider.

The AS3340 datasheet recommends (Vee-7.4) / 0.008, the CEM3340 datasheet recommends (Vee-7.2) / 0.008. For -12V, that's 575 ohms for the AS, 600 ohms for the CEM, so you should be good with your 620 ohms. The 3340 will run cooler if you give it a -5V supply instead of the -12, though.
In here somewhere is clearly my confusion— the current limiting resistor vs the voltage divider. Where is the voltage divider created in the datasheet schematic? Between the 600/620R resistor and... the trimmer?
There's no divider. The zener diode limits the negative voltage to -6.5 (CEM3340) or -7.4 (AS3340); the resistor limits the current to avoid damaging the zener.

Studioalethea
Wiggling with Experience
Posts: 265
Joined: Mon Dec 23, 2019 2:20 am
Location: Seattle

Re: CEM3340 Pin 3 Considerations

Post by Studioalethea » Wed Jul 29, 2020 5:54 pm

Borogove wrote:
Wed Jul 29, 2020 5:48 pm
Studioalethea wrote:
Wed Jul 29, 2020 4:39 pm
Borogove wrote:
Wed Jul 29, 2020 4:28 pm
socom93 wrote:
Wed Jul 29, 2020 2:39 pm
The datasheet says there is an internal6.5v zener diode. So no miscalculation they also give how to calculate the value of the limiting resistor.
Yep, Ree is a current limiter into the zener, not a resistive divider.

The AS3340 datasheet recommends (Vee-7.4) / 0.008, the CEM3340 datasheet recommends (Vee-7.2) / 0.008. For -12V, that's 575 ohms for the AS, 600 ohms for the CEM, so you should be good with your 620 ohms. The 3340 will run cooler if you give it a -5V supply instead of the -12, though.
In here somewhere is clearly my confusion— the current limiting resistor vs the voltage divider. Where is the voltage divider created in the datasheet schematic? Between the 600/620R resistor and... the trimmer?
There's no divider. The zener diode limits the negative voltage to -6.5 (CEM3340) or -7.4 (AS3340); the resistor limits the current to avoid damaging the zener.
Got it. Therefore, without the AS/CEM3340 in position, the socket of Pin 3 will indeed read -12V, while with the chip in place, will measure the above voltages (within tolerance)?

Soooo... where did we get that the 3340 prefers -5V on that pin? I am not seeing that in the data sheets, but there seems to be a lot of talk about it,

User avatar
infinitemachinery
Common Wiggler
Posts: 144
Joined: Sat Jan 05, 2019 11:01 am
Location: Ashland, Oregon
Contact:

Re: CEM3340 Pin 3 Considerations

Post by infinitemachinery » Wed Jul 29, 2020 6:16 pm

Studioalethea wrote:
Wed Jul 29, 2020 5:54 pm
Soooo... where did we get that the 3340 prefers -5V on that pin? I am not seeing that in the data sheets, but there seems to be a lot of talk about it,
It's not so much that the 3340 likes -5V on pin, it's the fact that you need a steady voltage on pin 3 in order to have consistent tuning. If pin 3 is supplied from the PSU mains then voltage can possibly vary depending on what other modules you are powering and other factors. Having a regulated supply, -5V, eliminates that problem and makes the performance of the 3340 more reliable.

Borogove
Learning to Wiggle
Posts: 37
Joined: Sun Feb 25, 2018 10:56 pm

Re: CEM3340 Pin 3 Considerations

Post by Borogove » Wed Jul 29, 2020 7:27 pm

Studioalethea wrote:
Wed Jul 29, 2020 5:54 pm

Got it. Therefore, without the AS/CEM3340 in position, the socket of Pin 3 will indeed read -12V, while with the chip in place, will measure the above voltages (within tolerance)?
Yep.

Studioalethea wrote:
Wed Jul 29, 2020 5:54 pm

Soooo... where did we get that the 3340 prefers -5V on that pin? I am not seeing that in the data sheets, but there seems to be a lot of talk about it,
It's in the AS3340 data sheet:

> To minimize self-heating and improve thermo-stability it is recommended to keep VEE = -5V (external power supply).

If you give it -12, the resistor and the zener are dissipating power as heat. If you have a 7905 regulator, the 7905 dissipates the heat and the 3340 stays cool. The 7905 has, depending on the package, a heat sink or at least a good sized thermal pad on it, and costs under a buck.
Last edited by Borogove on Wed Jul 29, 2020 7:29 pm, edited 2 times in total.

Borogove
Learning to Wiggle
Posts: 37
Joined: Sun Feb 25, 2018 10:56 pm

Re: CEM3340 Pin 3 Considerations

Post by Borogove » Wed Jul 29, 2020 7:28 pm

Note that all the 3340-using synth schematics that Electric Druid has collected (Roland, Sequential, and Moog) use -5V to pin 3 instead of the limiting resistor, so this appears to have been conventional wisdom before the AS3340 came along.

Since the datasheet does show the -15V and resistor solution, though, it probably isn't actually significantly bad for the chip.

Note that the 3310 ADSR and 3330 VCA also have the zener, and the 3320 VCF also has an "internal shunt regulator" providing -1.9V internally so also needs a limiting resistor on the V-. Since these parts were often used together, it would make a lot of sense to have a single voltage regulator upstream with a big fat heat sink far from the tuning-sensitive VCOs, and save a resistor per IC. That's an especially big win on a full featured poly -- a rev 3 Prophet 5 used 11x 3340s (LFO + 2 oscs per voice) and 10x 3310s (amp and filter envelope per voice), for example.

Studioalethea
Wiggling with Experience
Posts: 265
Joined: Mon Dec 23, 2019 2:20 am
Location: Seattle

Re: CEM3340 Pin 3 Considerations

Post by Studioalethea » Wed Jul 29, 2020 8:51 pm

Borogove wrote:
Wed Jul 29, 2020 7:28 pm
Note that all the 3340-using synth schematics that Electric Druid has collected (Roland, Sequential, and Moog) use -5V to pin 3 instead of the limiting resistor, so this appears to have been conventional wisdom before the AS3340 came along.

Since the datasheet does show the -15V and resistor solution, though, it probably isn't actually significantly bad for the chip.

Note that the 3310 ADSR and 3330 VCA also have the zener, and the 3320 VCF also has an "internal shunt regulator" providing -1.9V internally so also needs a limiting resistor on the V-. Since these parts were often used together, it would make a lot of sense to have a single voltage regulator upstream with a big fat heat sink far from the tuning-sensitive VCOs, and save a resistor per IC. That's an especially big win on a full featured poly -- a rev 3 Prophet 5 used 11x 3340s (LFO + 2 oscs per voice) and 10x 3310s (amp and filter envelope per voice), for example.
All good points. I forgot about the Electric Druid research. Seems pretty low risk to use a regulator. I think I might dive into that option after all, but thanks all for filling in my knowledge gaps.

User avatar
col
Wiggling with Experience
Posts: 478
Joined: Tue Feb 04, 2014 3:47 pm
Location: Melbourne

Re: CEM3340 Pin 3 Considerations

Post by col » Wed Jul 29, 2020 10:43 pm

Assuming you have read this thread: viewtopic.php?f=17&t=193367&start=175

Studioalethea
Wiggling with Experience
Posts: 265
Joined: Mon Dec 23, 2019 2:20 am
Location: Seattle

Re: CEM3340 Pin 3 Considerations

Post by Studioalethea » Thu Jul 30, 2020 3:26 pm

col wrote:
Wed Jul 29, 2020 10:43 pm
Assuming you have read this thread: viewtopic.php?f=17&t=193367&start=175
Thanks for directing me back to this! I had it bookmarked and looked earlier in my research but forgot to dive deeper into it with this specific question. There's a lot there! This place is such a gold mine... :hail:

Studioalethea
Wiggling with Experience
Posts: 265
Joined: Mon Dec 23, 2019 2:20 am
Location: Seattle

Re: CEM3340 Pin 3 Considerations

Post by Studioalethea » Thu Aug 06, 2020 11:30 am

I very easily installed the -5V regulator according to the data sheet, and it seems great so far! Everyone should do this! I was able to install it rather close to the power header and essentially run an additional power rail, and now all the heat dissipation is far from the CEM3340. Thanks all!

Post Reply

Return to “Music Tech DIY”