AS3340 14v pulse output

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

Moderators: Kent, Joe., analogdigital, infradead, lisa, parasitk, plord, sduck

User avatar
l3v3l6
Common Wiggler
Posts: 66
Joined: Sat Jun 09, 2018 10:03 am
Location: Ashland, Oregon

AS3340 14v pulse output

Post by l3v3l6 » Wed Dec 26, 2018 9:22 am

I'm working on a 3340 VCO board. In my oscope, saw and triangle waves look fine and match the datasheet spec. The pulse, on the other hand, looks okay in form and the PWM adjustment is working but the amplitude is at 14v peek rather than the 12v specified in the data sheet. What could be causing this? I'm driving the chip using +15v on pin 16 and -5v on pin 3.

User avatar
kassu
Wiggling with Experience
Posts: 279
Joined: Fri Dec 18, 2015 7:44 am
Location: Finland
Contact:

Post by kassu » Wed Dec 26, 2018 1:33 pm

The pulse output is open collector, and the amplitude depends on the load resistor and the output transistor. In my experience at least AS3340 has pretty large output swing, maybe the transistor is stronger than in the original CEM.

By the way, a pretty large load resistance such as 50k or more is recommended to limit how much PWM influences the supply and through that the tuning.
Synth DIY blog and circuits: http://kassu2000.blogspot.fi/

User avatar
l3v3l6
Common Wiggler
Posts: 66
Joined: Sat Jun 09, 2018 10:03 am
Location: Ashland, Oregon

Post by l3v3l6 » Wed Dec 26, 2018 1:55 pm

Thanks for the reply. I had put a 51K resistor on that output. I'll check to make sure that is actually the case.

alfa
Learning to Wiggle
Posts: 49
Joined: Tue Oct 24, 2017 10:40 pm
Location: Latvia
Contact:

Post by alfa » Wed Jan 02, 2019 12:58 am

Just for information :
Correct description of PWM output structure You can find
in datasheet of AS3340-HYB module:
http://www.alfarzpp.lv/eng/sc/AS3340-HYB.php

User avatar
devinw1
Super Deluxe Wiggler
Posts: 1073
Joined: Tue Aug 07, 2018 11:20 am
Location: Portland, OR
Contact:

Post by devinw1 » Wed Jan 02, 2019 11:46 am

Interestingly, the original Pro-One schematic shows a little picture of the square being 0 to +13V right out of the 3340... You're right the 3340 datasheet does show 12V though. I wonder if they've always been a little hot (not temperature wise, but the 12V peak sq being higher)
Last edited by devinw1 on Wed Jan 02, 2019 4:09 pm, edited 1 time in total.

alfa
Learning to Wiggle
Posts: 49
Joined: Tue Oct 24, 2017 10:40 pm
Location: Latvia
Contact:

Post by alfa » Wed Jan 02, 2019 1:52 pm

Certainly, if IC is hot it is not good. Among several good "steps":
1. Turn off internal zener ( approx. -7.4V) - use external negative
power supply -5- -6v. It will make internal heat dissipation
lower and some other things better.
2. Use positive +12v
3. Good solution - AS3345F - QFN package + correct
heatsink on pcb higly reduce thermal resistance
4 About tuning of AS3340/45 (by Rob Hordijk):
http://www.alfarzpp.lv/eng/sc/Tuning%20the%20AS3340.pdf
http://www.alfarzpp.lv/eng/sc/AS3340.php

User avatar
rolfdegen
Common Wiggler
Posts: 81
Joined: Sat Jan 12, 2019 6:48 am
Location: Wuppertal (germany)

Post by rolfdegen » Thu Jun 20, 2019 5:26 pm

Hallo

I want to develop a VCO with AS3340. AS3340 temperature behavior is not good. After switching on the mains voltage (stable + 10V pin 16 and -5V pin3) you have to wait 20min until frequency is stable.

Is that normal ?

Thank you for your answers. Greetings from old germany. Rolf :)

User avatar
tobb
Ultra Wiggler
Posts: 930
Joined: Sat Nov 26, 2011 5:41 am

Post by tobb » Thu Jun 20, 2019 6:46 pm

USE ONLY the CEM3340!!

alfa
Learning to Wiggle
Posts: 49
Joined: Tue Oct 24, 2017 10:40 pm
Location: Latvia
Contact:

Post by alfa » Thu Jun 20, 2019 11:23 pm

Dear Rolf
It looks you have problems in schematics. Please, re-send Your question
and schematics to: alfa@alfarzpp.lv

User avatar
rolfdegen
Common Wiggler
Posts: 81
Joined: Sat Jan 12, 2019 6:48 am
Location: Wuppertal (germany)

Post by rolfdegen » Fri Jun 21, 2019 2:01 am

@alfa

Thanks for your answer. My schematics is coming soon..

User avatar
Grumble
Super Deluxe Wiggler
Posts: 1552
Joined: Thu Dec 17, 2015 5:24 am
Location: U aint much if U aint Dutch
Contact:

Post by Grumble » Fri Jun 21, 2019 2:45 am

tobb wrote:USE ONLY the CEM3340!!
Why is that? For repair of synths using the CEM3340 I agree, but new designs....
Electric Druid wrote:The AS3340 chip is Alfa’s modern equivalent of the famous CEM3340 voltage-controlled oscillator.

It provides Ramp, Pulse/Square, and Triangle outputs, has inputs for oscillator sync, is temperature compensated and stable, and provides excellent 1V/Oct tracking. In short, it’s everything you need from an analog oscillator, on a chip. There hasn’t been a better design than the 3340.

Please note that Alfa recommend a 51K resistor on the pulse output in place of the 10K resistor that the CEM3340 used. There seem to be some subtle differences between the Alfa chips and CEM originals, particularly with respect to pulse width. We’ve had one or two reports of difficulties replacing dead/missing CEM3340s with AS3340s. For this reason, if you’re repairing a vintage synth, we recommend using CEM3340 Rev.Gs rather than replacing with Alfa AS3340s.

User avatar
rolfdegen
Common Wiggler
Posts: 81
Joined: Sat Jan 12, 2019 6:48 am
Location: Wuppertal (germany)

Post by rolfdegen » Fri Jun 21, 2019 5:42 am

How long time after power on does it take for your oscillator (AS3340) to stabilize?

alfa
Learning to Wiggle
Posts: 49
Joined: Tue Oct 24, 2017 10:40 pm
Location: Latvia
Contact:

Post by alfa » Fri Jun 21, 2019 5:54 am

Dear Rolf
Depends from schematics. If incorrect schematic or trimming procedure - unpredictable time. We need just part of Your schematics.

User avatar
rolfdegen
Common Wiggler
Posts: 81
Joined: Sat Jan 12, 2019 6:48 am
Location: Wuppertal (germany)

Post by rolfdegen » Fri Jun 21, 2019 6:41 am

Here my schematics. None Freq. trimmer for better test.

Power on frequency ist 700Hz. The stabel frequency after 20min is 800Hz.

I have test with two AS3340 Typ Nr. is AS3340 1808

Rt is normal metall resistor with 5,6k 1%

Image

Thanks an greetings. Rolf

alfa
Learning to Wiggle
Posts: 49
Joined: Tue Oct 24, 2017 10:40 pm
Location: Latvia
Contact:

Post by alfa » Fri Jun 21, 2019 8:48 am

Dear Rolf

You must trim temperature coefficient of AS3340. For these purpose
pin1 and pin 2 are used. Good procedure is described in Rob Hordijk article which is placed on our web-site:
http://www.alfarzpp.lv/eng/sc/Tuning%20the%20AS3340.pdf
http://www.alfarzpp.lv/eng/sc/AS3340.php

Main simple thing - through pin1 and pin 2 must flow equal current approximately 120 uA.
So, first step - You must use Rt resistor which gives You such current and after that trim current on pin1 with Rz.
Better solution of these place described in Rob Hordijk article.
Please, try to follow described in article instructions and solution.

User avatar
rolfdegen
Common Wiggler
Posts: 81
Joined: Sat Jan 12, 2019 6:48 am
Location: Wuppertal (germany)

Post by rolfdegen » Fri Jun 21, 2019 9:15 am

I have test with +12V to VCC and -12V in series with a 820R resistor to VEE. Its same problem. A lot of warmup time.

User avatar
rolfdegen
Common Wiggler
Posts: 81
Joined: Sat Jan 12, 2019 6:48 am
Location: Wuppertal (germany)

Post by rolfdegen » Fri Jun 21, 2019 9:35 am

I have test yesterday after the description of Rob. Same Problem. Power on and long warmup time.

Now i have a new eurorack case from doepfer with another power supply.

I will test with this..

Thanks a lot alfa :))

User avatar
OB1
Common Wiggler
Posts: 184
Joined: Fri Jan 12, 2018 7:11 am
Location: Bristol, UK
Contact:

Post by OB1 » Fri Jun 21, 2019 9:46 am

rolfdegen wrote:you have to wait 20min until frequency is stable.

Is that normal ?
I have never even bothered to attempt to tune ANY analogue oscillator until it's been turned on for at least 20 minutes.

User avatar
infinitemachinery
Common Wiggler
Posts: 96
Joined: Sat Jan 05, 2019 11:01 am
Location: Ashland, Oregon
Contact:

Post by infinitemachinery » Fri Jun 21, 2019 9:55 am

OB1 wrote:
rolfdegen wrote:you have to wait 20min until frequency is stable.

Is that normal ?
I have never even bothered to attempt to tune ANY analogue oscillator until it's been turned on for at least 20 minutes.

Same. My Moog Voyager needs at least 30 mins.

User avatar
devinw1
Super Deluxe Wiggler
Posts: 1073
Joined: Tue Aug 07, 2018 11:20 am
Location: Portland, OR
Contact:

Post by devinw1 » Fri Jun 21, 2019 10:43 am

I've found the CEM3340s (in my Pro-One clones anyway) need about 5 minutes to settle in. At least that's until it's in tune to my ear to the Rhodes it sits atop. It's always sharp by quite a bit on first turn on, but comes back exactly to where it was and totally playable and in tune with the piano after 5 minutes.

alfa
Learning to Wiggle
Posts: 49
Joined: Tue Oct 24, 2017 10:40 pm
Location: Latvia
Contact:

Post by alfa » Fri Jun 21, 2019 12:07 pm

Dear Rolf

Sorry, why for -12V supply you used 820 Ohm? It is out of specification,
and IC can't work correctly.
Please, follow datasheet - recommendation how to choose these resistor.
Another moment - using internal zener - is not the best practice.
Additional heating , need more compensation and heating time/equilibrium.

User avatar
rolfdegen
Common Wiggler
Posts: 81
Joined: Sat Jan 12, 2019 6:48 am
Location: Wuppertal (germany)

Post by rolfdegen » Sat Jun 22, 2019 6:21 am

Hallo alfa..

Thanks for your tip: I change my Vee-supply and make it directly with -5V from 79L05. VCC is +12V from my Eurorack. I change the Tempco Pin 1+2 and Pin 14+15 with Robs example.

My main question is: Need the AS3340 a warm-up time after power on ?

Thanks for your help. Greetings Rolf

alfa
Learning to Wiggle
Posts: 49
Joined: Tue Oct 24, 2017 10:40 pm
Location: Latvia
Contact:

Post by alfa » Sat Jun 22, 2019 6:47 am

Dear Rolf
Good news!
Somerimes , using 79l05, diode is placed in gnd pin - raising Vee to approx. -5.7 v.
Using external -Vee supply ( turning off internal zener) minimize additional heating dissipation inside.
Yes, certainly, every IC has warm-up. We don't define it.
If you trimmed tempco - you minized these time and made vco more stable.
Size of solder pads and other construction moments have big
influence on heat dissipation/equlibrium. Don't forget, that temperature
stability of resistors connected to pin 1 & 2 has also influence on result.

User avatar
rolfdegen
Common Wiggler
Posts: 81
Joined: Sat Jan 12, 2019 6:48 am
Location: Wuppertal (germany)

Post by rolfdegen » Sat Jun 22, 2019 7:05 pm

The diode has a temp drift !?

nigel
Veteran Wiggler
Posts: 603
Joined: Wed Jan 16, 2013 6:49 am
Location: Melbourne, Australia

Post by nigel » Sat Jun 22, 2019 7:36 pm

rolfdegen wrote:The diode has a temp drift !?
Welcome to the wonderful world of electronics. Everything has a temperature drift.

Post Reply

Return to “Music Tech DIY”