Hi guys
NJKMonty has been grafting away with Eagle designing a layout for a CE-1 clone.
Heres the schematic (http://i999.photobucket.com/albums/af111/njkmonty/ScreenHunter_07%20Apr.%2001%2012.27_zpsxy2tvexj.jpg) we are using.
Due to the size of the of the board and our view to make it in a 1590bb we're struggling with the power section.
In a nutshell, we are trying to get a +/-14v supply from 9v.
I was looking at doing something like this (http://tagboardeffects.blogspot.com/2014/01/bipolar-voltage-converter.html) but with different diodes to get us closer to +-14v.
(http://3.bp.blogspot.com/-TsErabpIPyU/UssKrgKIsxI/AAAAAAAAF3M/6xl3AguLyGY/s1600/Bipolar+Voltage+Converter.png)
But I'm sure theres a better way.
Any suggestions would be more than welcome :)
one fellow member here suggested something like this however i think they only handle low amps?
https://www.rapidonline.com/mornsun-wra0515s-3wr2-3w-dual-o-p-sip-regulated-isolated-dc-converter-15v-dc-84-1561
This will bring you to +14.5/-13.7 V:
(http://i64.tinypic.com/294m49v.jpg)
Or you can use Mornsun A0915S-1WR2, it is 2W Fixed Input Isolated and Unregulated Dual Output DC/DC Converter.
http://www.acalbfi.com/uk/Power-supplies/DC-DC/PCB-mounted/p/2W-Fixed-Input-Isolated-and-Unregulated-Dual-Output-DC-DC-Converter/0000005BDY (http://www.acalbfi.com/uk/Power-supplies/DC-DC/PCB-mounted/p/2W-Fixed-Input-Isolated-and-Unregulated-Dual-Output-DC-DC-Converter/0000005BDY)
This efficient DC/DC converter runs in the MHz so....it's way out of human audio range
The output voltage can be changed with a couple of resistors.
https://obsoletetechnology.wordpress.com/projects/studio-electronics/dc-dc-bipolar-power-supply-for-effect-pedals/
Cheers guys. That's excellent.
Thanks as always Larry
would those smd capacitors voltage rating be adequate?
Quote from: njkmonty on December 03, 2017, 06:03:20 PM
would those smd capacitors voltage rating be adequate?
If you are referring to the supply that I posted....yes.
You would be surprised at the voltage ratings of tiny SMD parts.
Miniature ceramic caps are rated for 1000V!
I have a BOM of parts that I sourced from Digikey...I can PM you the list if you'd like.
Thanks heaps
You had me at CE-1. :icon_cool:
Any plans for public boards?
Yep. Thats the plan.
We're looking at getting some boards made and then selling them with an MN3002.
Aussie Nick (Njkmonty) has done most of the work (actually pretty much all of the work), and from what I've seen its really coming on!
yeah its point to point but turrets arent cheap, so not sure what it will cost
A point-to-point Ce-1 clone on turrets? Well this I gotta see.
I currently have in my posession a Multivox CB-1 Chorus Box, which I've read is an exact CE-1 clone (not sure of that, but it's definitely close). I fixed it for a buddy who lives in a neighboring state, so it's here with me until he comes to town and gets it, along with a few other items too big to ship. If you hit a snag and need voltages, I'd be glad to fire it up and try to help, assuming it's still here. Can also take some gut pics, quality limited by my crappy camera, if that might help you in any way.
sure! why not
turrets was me taking the piss out of my limited eagle cad abilities in making a good small layout!
Here ya go, let me know if you want anything more specific or zoomed.
(https://s2.postimg.org/el0ls7y0l/DSCF2340.jpg) (https://postimg.org/image/el0ls7y0l/)
(https://s2.postimg.org/ng1g2upet/DSCF2341.jpg) (https://postimg.org/image/ng1g2upet/)
(https://s2.postimg.org/k96wj6myd/DSCF2342.jpg) (https://postimg.org/image/k96wj6myd/)
(https://s2.postimg.org/n3a1wnrph/DSCF2343.jpg) (https://postimg.org/image/n3a1wnrph/)
(https://s2.postimg.org/hf3r5qshx/DSCF2344.jpg) (https://postimg.org/image/hf3r5qshx/)
(https://s2.postimg.org/aon9wbshh/DSCF2345.jpg) (https://postimg.org/image/aon9wbshh/)
(https://s2.postimg.org/z55fqslid/DSCF2346.jpg) (https://postimg.org/image/z55fqslid/)
(https://s2.postimg.org/i4mji2qgl/DSCF2347.jpg) (https://postimg.org/image/i4mji2qgl/)
(https://s2.postimg.org/mdr9kayvp/DSCF2348.jpg) (https://postimg.org/image/mdr9kayvp/)
I'm not generally a big fan of chorus, but this thing does sound really nice. Think "Feeling Gravity's Pull" by REM. Gotta keep the speed pretty low (here labeled "intensity" for some reason). Best below 1Hz I'd say, otherwise it sounds more like your guitar's going out of tune rhythmically, make you seasick. The vibrato function is pretty useless IMO, doesn't sound very good, but it's neat that it has an independent speed control.
that looks like the same board as the original!!!
fantastic
Wow, holy crap!
https://medias.audiofanzine.com/images/normal/boss-ce-1-chorus-ensemble-672952.jpg (https://medias.audiofanzine.com/images/normal/boss-ce-1-chorus-ensemble-672952.jpg)
Exactly the same board, stuffed with different brands of components. I guess Roland sold the PCBs to Multivox. Both MIJ, but unrelated companies AFAIK.
About 20 years ago, another buddy of mine had a Multivox tape echo, with a built-in chorus. That thing sounded amazing, otherworldly. It was cool just to watch, the tape wasn't on reels but just snaked around in a shallow chamber on top that had a smoked plexiglass lid. I'd love to find one, but I wouldn't want to have to work on it.
Hey Armdnrdy,, if we implement the smd bipolar power supply into the ce 1 circuit, would we need to use all the 220uf and 47uf power capacitors? or do you think what you suggested would not need any more capacitor filtering?
in pic there are 2 examples, would the bottom be ok, diodes are just to drop voltage
(https://s5.postimg.org/qhkw4rxzb/Screen_Hunter_03_Dec._07_18.44.jpg)
We dont need to use diodes to get down to + and - 14v, we need to use the values of R1 and R2 to get us the 14v (as per the webpage linked by armdnrdy). We just need to also get the +/-13v after that
Quote from: njkmonty on December 07, 2017, 03:46:43 AM
Hey Armdnrdy,, if we implement the smd bipolar power supply into the ce 1 circuit, would we need to use all the 220uf and 47uf power capacitors? or do you think what you suggested would not need any more capacitor filtering?
in pic there are 2 examples, would the bottom be ok, diodes are just to drop voltage
I would say....additional power supply filtering isn't going to hurt anything but...
I don't see the need for the size used in the original.
The original circuit used larger caps to smooth the PS after conversion from AC to DC.
I would use 100µf after the 3467 supply, and 47µfs for the 13V.
Hi Guys
Sorry to bring this back up but we feel like we just haven't got this licked.
I love the link from Larry(Armdnrdy) (https://obsoletetechnology.wordpress.com/projects/studio-electronics/dc-dc-bipolar-power-supply-for-effect-pedals/) but, as we've had a lot of interest in this project, me and Aussie Nick are planning on making up a batch of boards and are worried that the smd will put off a lot of people. It puts me off. I ordered some smd chips and my wife nearly threw them in the bin because she reckoned rs components had delivered 2 empty bags.
I've been searching the internet to find a DIP equivalent to the LT3467. I'm sure there must be something but I'm probably not searching correctly. It would be nice to do it with one chip.
We've also been talking about using 2 chips and a regulator.
IC1 to double then using a regulator (like an LM317 (https://www.onsemi.com/pub/Collateral/LM317-D.PDF) to get 14v), then inverting for the -14v rail.
Or perhaps 2 chips and 2 regulators which might be better for the current draw. IC1 doubles the 9v and then we regulate, IC2 doubles the 9v and inverts then we use negative voltage regulator (https://www.onsemi.com/pub/Collateral/LM337-D.PDF) for the -14v.
Any ideas will be more than welcome. Board space is tight as its such a big circuit so the fewer components the better. We've got something that works but are worried we are doing it like idiots! :icon_biggrin:
There is another option of making the boards with the smd chip already onboard but that brings in some logistical challenges.
have a look at the good old MC34063.
Thats excellent ducky. Cheers buddy.
I take it I'll need to use 2 to get the +14 and the -14v or is there a way it can do both with one IC like the LT3467?
Hmm
I keep coming across this image (or variations) of it from places like here. (https://electronics.stackexchange.com/questions/164038/what-type-topology-of-smps-is-this)
(https://i.stack.imgur.com/RHnNv.png)
Only thing I'm not sure about is that transformer. What is that? A flyback converter apparently. Dont know if its worth doing it that way or just use two MC34063s. :-\
Why not using LT1054? Proven and small.
Sorry Ranko, thanks for your schematic. I dont know why we have ignored it, I think I got a bit obsessed with the LT3467 alternative route. We were also concerned about the max current from the LT1054. From Aussie Nicks tests on his working clone the current on each rail seemed high.
I'll have a word with Nick and see if he fancies breadboarding your schematic.
I'm crook at present but will advise my results
Bipolar from lt1054 tested perfect until I connected to circuit the volts dropped too low +6. And -4v
However I've managed to just cram in a dual lt1054 with a to220 regulator and it just fits ( regulator mounted on pot side of board ) the dual surge pump idea worked perfectly in my clone circuit. And was as quiet as my 14vac transformer version
I have. No probs at all with two lt1054's. However I've just never seen anyone do it that way bedore
It's also. ISnt smd And a common chip pedal makers source?!
The non-DIY path is DC-DC converter. Used to be used by the millions on Ethernet cards; also widely around telephony cards (all dead markets). Cost you 20 bucks, but may be cheaper than throwing-out bags of "pepper grains".
Digikey Search link {may not work for you} (https://www.digikey.com/products/en/power-supplies-board-mount/dc-dc-converters/922?FV=8f40095%2C8f40096%2C8f40097%2C8f4009e%2C8f400a8%2C8f400b2%2C8f40013%2C8f40014%2C8f40016%2C8f40017%2C8f40018%2C8f40019%2C8f40022%2C8f40023%2C8f40024%2C8f40028%2C8f4002e%2C8f40032%2C8f40034%2C8f40042%2C8f4004d%2C8f40050%2C8f40052%2C16fc000b%2C16fc0014%2C16fc0027%2C16fc0028%2C16fc0057%2C16fc0009%2C1f140000%2Cii2%7C2211%2Cffe0039a%2C17d4002c&mnonly=0&ColumnSort=1000011&page=1&stock=1&pbfree=0&rohs=0&cad=0&datasheet=0&nstock=0&photo=0&nonrohs=0&newproducts=0&k=voltage+converter&quantity=&ptm=0&fid=0&pageSize=25&pkeyword=voltage+converter)
RSO-1212DZ Recom Power CONV DC/DC 1W +/-12VOUT SIP $18.22
IP1212S XP Power DC/DC CONVERTER +/-12V 3W $19.95
ISU0205D12 XP Power DC DC CONVERTER +/-12V 2W $21.00
IEQ0512D12 XP Power 5W ISOLATED DC-DC CONVERTER, 4:1 $23.00
ISU0205D12 XP Power DC DC CONVERTER +/-12V 2W $25.00
Connections:
(https://s18.postimg.org/ikq8nux85/RECOM-_RSO.gif) (https://postimg.org/image/ikq8nux85/)
You can also try with this: https://www.aliexpress.com/item/dc-dc-converter-power-module-5v-step-up-to-positive-and-negative15v-dual-output-isolated-dc/1935768851.html
It is cheap (5 pcs for 12 USD) and convert 5V DC to ±15V DC. They can probably offer other options - 9V input for example, but it is not easy to search and find on this site. Drawback is that can supply only 67mA.
If you need more (200mA), try with this: https://www.aliexpress.com/item/2-1-wide-voltage-3w-0-2A-power-supply-module-12v-to-15v-dual-isolated-regulated/2027275014.html
Or you may try using XL6009, looks quite simple and works on 400kHz!
https://lib.chipdip.ru/164/DOC001164799.pdf
i told them about the dc dc converters but they didn't listen
i valued your input, however sourcing the right one with out having to spend $$ on a minimal order has been an issue,, i was trying to work around with common parts availble to me, and parts used frequently in diy community projects,, if that all fails then look down that road.
ive got it to work well and have enough room to fit it all on my board layout, I cant see whats wrong with using dual lt1054's? Cost less than 14 - 20us dollars?
its just the part cost and the board size isnt going to change?
also if we end up getting a working project, we would like to make it so parts used are as easier and cost effective in sourcing
this is sort of what we have so far....
the 6 diodes in series are there with the option of jumpering them to get the desired voltage pending own tests
(https://s5.postimg.org/4sgminzl3/Screen_Hunter_02_Jan._06_17.52.jpg)
(https://s5.postimg.org/l3gqez4d3/Screen_Hunter_01_Jan._06_17.51.jpg)
where will the 3 jacks go tho?
hows your layout going? did you end up finding any bbd chips?
Looks like you need to ground the junction of C47 and C48.
Also, you should find out how much current the circuit will draw since charge pump voltages will sag significantly as you load it down. You might need to take this into account.
Best regards, Jack
Not started yet, I have 6 Panasonic MN3002s in my stash.
i tested my first one and from memory each rail was drawing between 30-60 ma
yes i saw that schem,atic correction ive since corrected a few values etc and finished , was just showing which way we went , however you have good eyes!
6 chips ! impressive :)
These look like they would give you +-15 V at 100mA per rail, and it's through hole. No idea of cost though
https://www.tracopower.com/products/tmr3wie.pdf
cheers, i like the look at all these power ic packages , but i cant ever seem to find anyone who sells them without having to spend of $100 on an order!
https://au.mouser.com/ProductDetail/TRACO-Power/TMR-3-1223WIE/?qs=ckJk83FOD0VpG70EJMzayg==
$24 !
i found some other ones on there that looked good but only handle +-33ma and were under $4
https://au.mouser.com/productdetail/mean-well/dpu01m-15?qs=sGAEpiMZZMvGsmoEFRKS8Koqt8Pjkl39mGzvujnWEEkR3klXv%2Fo9dg%3D%3D
i think i might stay with the dual lt1054 ($3 each) and i use them in builds regularly) however i have one more single lt1054 layout to try!
however retro sonic cleary uses a dc dc converter on their boards (they also use the Tl072 aswell)
(https://s5.postimg.org/807bqd7dz/Retrosonic_Chorus_1.jpg)
also just to confirm all single lt1054 chip ideas haven't worked under load,
this one below displayed all the specified voltages until once connected the volts dropped by half on both rails
(http://i64.tinypic.com/294m49v.jpg)
this is the size of the dual lt1054 (minus the drop down diodes ) (on my other board)
so i feel at far less than $24 doesnt take up too many parts?
im not sure if i can get a smaller package 15v regulator aswell ie T0-92 instead of a T0-220 (the big one fits on my board but the smaller package only handles 100ma and i dont have any on me?
any suggestions regarding this too?
(https://s5.postimg.org/l4cw39wxj/image1.png)
78L15 is in TO-92 case, but it can handle only up to 100mA.
Regarding circuit with one LT1054, I believe that maximum current that you can expect for stable voltage is 35mA on both rails.
i might source some 78L15's and give it a go
if one rail is 40ma and the other is -40ma does that mean its a total of 80ma current draw?
also if hypothetically, (will measure it again) my peak current drawer is say 82ma, is a 100ma max component have adequate head room?
But if you already use 2 pcs of LT1054, you don't need 7815 at all, you can arrange them like this:
(http://i63.tinypic.com/fjp5hv.jpg)
LTspice shows that output is +15.4/-13.4 V for 100mA load on both rails, which is maximum output of a single LT1054.
i see and just tweak load resistors to taste?
ok ive finished breadboarding the last dual LT1054 schematic from Rankot
I actually got the lfo to work (i didnt plug the ce 1 sonically , as ive found if the voltages are in the correct range the throbing lfo etc works good)
however i had to make some adjustments
in order to get the circuit providing as close to 14v on both rails, i had to omit Rload2 as without it it was only producing -12.5v (under load)
Rload1 i found that a 200 r resistor worked best ,
however .....
the Rload1 resistor was getting extremely hot!
i put in a 1/2 w resistor and it was getting very hot too
i disconnected to prevent any issues, and came to the conclusion that even though the led was throbbing the volatges were only ideal on the Positive rail, and the hot resistor really worried me that it was going to fail if left on,
and that
my original dual lt1054 plus regulator worked fine without these issues
any comments or thoughts would be happy to hear!
Quote from: njkmonty on January 07, 2018, 05:25:10 AM
in order to get the circuit providing as close to 14v on both rails, i had to omit Rload2 as without it it was only producing -12.5v (under load)
Rload1 i found that a 200 r resistor worked best ,
however .....
the Rload1 resistor was getting extremely hot!
i put in a 1/2 w resistor and it was getting very hot too
Rload is not a component you need to include, it is a stand-in for the circuit in simulation, a resistor that will draw a current approximately the same as the circuit will. This simplifies simulation so you don't have to model the entire circuit and bog down your computer with simulating the entire thing (because all we want to know is that the power supply works, not that the CE1 circuit works).
oh!
(embarrased!)
;D
Of course, remove Rload resistors from breadboard!
And without them, most probably the first one schematic will work (using only one LT1054).
ok here is results
using Rankot's single lt1054 (minus the Rload) didn't work , it measure properly but under load dropped dramatically
using Rankot's Dual lt1054 (minus Rlaods 1 & 2 :) ) provided greater than 15v on the + rail and only 12.4 on negative, when connected , it was enough to make the led come on but wouldn't throb with lfo
im happy to pursue my other 2 x surge pump layout plus regulator which has tested well, my only questions now are, ..
since ive mashed together different layouts i was wondering if a few components may be removed or reduced in value?
particularly the 15v zener diode
and 5 x 220uf caps?
surely some of these can be reduced or removed?
(https://s5.postimg.org/fbm4o9zs7/Screen_Hunter_03_Jan._08_23.05dsfsdf.jpg)
The 15v zener is over-voltage protection for IC7. If you know that you will never connect more than 15v to it, then leave out the zener.
You need the 220u caps.
Diodes D29 - D32 are not needed and are part of the problem.
The input voltage from IC7 will sag and is too close to the low limit on the LM7815 for me, so I would use a low dropout regulator like the LM2940. You might get by with the LM7815 but it is close.
Delete D27 too.
(http://www.muzique.com/misc/regulators1.gif)
regards, Jack
Thank you for replying!
most schematics involving the zener ive come across have only come from madbean layouts!
so i wasn't sure how important it may be! ive got some room on my layout in that area , so i will just leave it in
i currently have the layout shown previously working on my clone, using the regular regulator!, but will buy some of the LM2940ct-15 soon
however regarding the 3 diodes on each rail after ic 8 are more like pads to add and remove diodes for taste
to get mine in the working 14v i need 3 diodes in series on the + side, and only 2 on the negative.
unless there is another way (with minimal, easily sourced parts to drop that .75-1 volt from 15 down to 14 , then i was going to stick with the diodes as they seem to be working!
something like this?
(https://s5.postimg.org/7h7qxvt1j/Screen_Hunter_05_Jan._09_21.29.jpg)
i went out and remove some of the items and found now after updating it , it seems good after removing some of the advised diodes
currently 1 diode on the +
and no diodes on the negative rail (once load is connected , seems good! )
i also increased all 5 caps to 220uf
Hi I had success powering that4301 compressor with this inexpencive dcdc converter, its a 3W dual 15V. And output voltage can be adjusted with resistors. It might work for your circuit, here it is on ebay :
https://www.ebay.co.uk/itm/15V-Positive-Negative-Dual-DC-DC-Step-up-Boost-Converter-Module-3-3V-5V-12V/172763714588?hash=item283985141c:g:1~YAAOSwzx9Zw6Mw
(https://i.ebayimg.com/images/g/1~YAAOSwzx9Zw6Mw/s-l500.jpg)
That seems nice and clean. Thanks
How do you adjust the voltages with resistors?
the only chip I could think of is this XL6007
http://www.ksmcu.com/pdf/XL6007%20datasheet.pdf
there is two resistors to adjust feedback, look in datasheet. I did not need to adjust voltage for my application.
or it is possibility to use 14V zeners.
BTW the chip i figuring out is from here also on ebay its can be seen on images :
https://www.ebay.co.uk/itm/Positive-Negative-Dual-Output-Power-DC-Step-up-Boost-Converter-Module-5V-15V-VD/253246459737?ssPageName=STRK%3AMEBIDX%3AIT&var=552368949139&_trksid=p2060353.m1438.l2649
(https://i.ebayimg.com/images/g/s7wAAOSwMQBaABQB/s-l1600.jpg)
I can't understand why you make 18V with LT1054, then regulate that down to 15V, and then use that same voltage for inverting? It means that you are limited to total maximum output of first LT1054, which is 100mA, and your circuit obviously doesn't use more.
You may try this circuit, too:
(http://i64.tinypic.com/2rf4gfc.jpg)
If you experience voltage drop under load, you may try to remove regulation part of a circuit (transistors and zeners).
Quote from: rankot on January 09, 2018, 12:20:40 PM
I can't understand why you make 18V with LT1054, then regulate that down to 15V, and then use that same voltage for inverting? It means that you are limited to total maximum output of first LT1054, which is 100mA, and your circuit obviously doesn't use more.
You may try this circuit, too:
(http://i64.tinypic.com/2rf4gfc.jpg)
If you experience voltage drop under load, you may try to remove regulation part of a circuit (transistors and zeners).
I do agree with you. I think it would be best to separate the two LT1054, since when you cascade them not only you are limited to the first IC 100mA, but you are adding a load of some mA to the first IC coming from the second...
It seems really strange that your schematic wasn't working properly with two separate LT1054... maybe the negative doubler configuration doesn't provide enough voltage on its own? Is it worth maybe a second try? I have found as well a datasheet which shows how to get +2*9v and -2*9V with two LT1054... this should work I think!
(https://s10.postimg.org/m88y7jvit/Schermata_2018-01-09_alle_19.27.03.png) (https://postimg.org/image/m88y7jvit/)
Anyway, if you guys really want to be limited to the 100mA from the first IC, there is a Bipolar Supply doubler schematic floating around. I can try to build it, but I wouldn't know how to test it under load.
(https://s10.postimg.org/eppsz2whx/Schermata_2018-01-09_alle_19.12.10.png) (https://postimg.org/image/eppsz2whx/)
PS: if the CE-1 mA request is under 100mA I wouldn't think about it too much and use the first working schematic :icon_twisted:
XL6007 is actually very cheap - you can buy 10 for some 2 USD.
https://www.aliexpress.com/item/5pcs-lot-XL6007E1-SOP-8/32820416393.html
I'm very interested in using the MC34063A (https://www.sparkfun.com/datasheets/IC/MC34063A.pdf). Its DIP, can invert and step up and ticks all the boxes.
I found this calculator (http://www.nomad.ee/micros/mc34063a/) online which allows you to key in your input voltage, desired output voltage etc and then it gives you the schematic and BOM.
(https://docs.google.com/uc?id=10cmMlySn89gquY-R3W5d-jwjnxWMLdB3)
I'm just a bit confused what I should be setting the Vripple and Fmin parameters to, if we intend to use it for the CE-1? I take it Vripple is output ripple, but I dont know what that is! ???
Anyone willing to give me an explanation and some guidance, it would be most appreciated
I've tried MC34063A in one of my pedals as a high voltage generator (35V) and it was too noisy. I replaced it with Schmitt trigger step up which was very nice.
Fmin is oscillating frequency of MC34063A, it shall be at least 25kHz for audio applications (but it may be safer to push it to 50).
Vripple is acceptable voltage ripple of your output voltage.
ive been playing around with many layouts and could not get 2 x surge pumps under load to give +/- 14v (in parallel mode) the negative rail would always only get to about -12.5
so i had a crack at 3 chips!
the following schematic i used got both rails on +/- 14 under load and seem to work fine,
i didnt hear any noise when playing the chorus, and appeared just the same as my 2 x lt1054 in series layout that worked for me,
i guess this allows each surge pump to handle the amps better?
also since the chips max can handle is 100ma, then i can use a smaller regulator T0-92 package, which is also 100ma rated.
i measured the Dc amps being used on circuit straight out of the (9V Boss power supply and this chorus seems to draw about 170ma?
would love to hear others thoughts / comments
(https://s5.postimg.org/4j2ifketj/Screen_Hunter_06_Jan._12_14.57.jpg)
i also purchased some of the following to test aswell ......
https://www.ebay.co.uk/itm/Positive-Negative-Dual-Output-Power-DC-Step-up-Boost-Converter-Module-5V-15V-VD/253246459737?ssPageName=STRK%3AMEBIDX%3AIT&var=552368949139&_trksid=p2060353.m1438.l2649
Quote from: lietuvis on January 09, 2018, 09:16:00 AM
Hi I had success powering that4301 compressor with this inexpencive dcdc converter, its a 3W dual 15V. And output voltage can be adjusted with resistors. It might work for your circuit, here it is on ebay :
https://www.ebay.co.uk/itm/15V-Positive-Negative-Dual-DC-DC-Step-up-Boost-Converter-Module-3-3V-5V-12V/172763714588?hash=item283985141c:g:1~YAAOSwzx9Zw6Mw
(https://i.ebayimg.com/images/g/1~YAAOSwzx9Zw6Mw/s-l500.jpg)
Unless you know exactly which IC you are dealing with here....I wouldn't chance it.
I punched on the ebay link and read something about 20ma.
There is no way of knowing what frequency this supply runs at either.
Most things aren't bothered by audio frequencies...but the things we build...are.
I know that the guys are searching for a through hole component that fits all of their needs but...the market does not support that.
I had searched for quite a while and came up empty.
The LT3467 DC/DC converter runs at an ample frequency, supplies ample current, and supplies the proper voltage.
Downfall...it's SMD. Have you considered using the LT3467 with other through hole components to make up the supply?
If one's soldering skills lacks the ability to solder a single six leg, surface mount IC...they might want to consider a different hobby. ;)
smd doesnt bother me,, its more that im trying to cater for those who can't!
im just trying to figure out what parts i can use !
do you think it will work?
the LT3467? under load? as everything else ive tried works great until i connect it !
also do you recommend the LT3467 or LT3467A
also you mentioned this works upto aboout 80ma each rail.
im not sure the correct way of measuring the chorus's amp usuage, , or do i measure each rail after it goes through power section?
i just quickly measured the amps on each rail (making sure it ran through my dmm and had it on DC amps!
results
negative rail 60-80ma (cycling with the lfo )
positive rail 13-14 ma (cycling with the lfo )
do you think it will work with these results?
after coming across this on the data sheet i feel a little more positive about your idea
(https://s5.postimg.org/qrb368frr/Screen_Hunter_07_Jan._12_17.36.jpg)
my dual lt1054 version in series works
should i just stick with that?
if your dual LT1054 works and you have enough space, I guess just stick with that. I don't see the point in searching other solutions that might not work as well as the LT1054. As far as I know it's one of the few converters which doesn't give any "audio" issue.. anyway in order to measure the draw you should simply put the multimeter in between the +15 supply and the effect for the + draw and in between the -15 supply and the effect for the - draw.
If you did like so, you should have enough reliable results, even if it sounds a little strange to me that the two + and - draws differ so much... anyway, it should be no issue as long as it's under 100mA
I am really unpatient to see what you come up with, so I would just guess stick with the LT if you tested it under several supplies and situations and it turned out to work well.
Yes that's how I measured. It. With my fluke dmm
Quote from: njkmonty on January 12, 2018, 01:51:16 AM
also do you recommend the LT3467 or LT3467A
Either one should put you well above audio range and the chance of heterodyning issues with the CE-1 mutivibrator.
Quote from: njkmonty on January 12, 2018, 01:51:16 AM
also you mentioned this works upto aboout 80ma each rail.
I believe that was stated in the page that I previously linked, Obsolete Technology. The author stated that he has used the LT3467 successfully on circuits requiring 80ma.
Quote from: njkmonty on January 12, 2018, 01:51:16 AM
negative rail 60-80ma (cycling with the lfo )
positive rail 13-14 ma (cycling with the lfo )
These readings seem very lopsided. It doesn't seem that the circuit would be designed so dependent on the negative rail. There might be issues with your build.
Now...the reasons I wouldn't suggest the LT1054 for this build.
Let me start by saying that I regularly use the LT1054 as my go to IC to create a bipolar voltage in circuits that don't require too much current.
When I moved to circuits with a higher current draw...I had to move to find a different solution. The LT3467 is one of the solutions....others include center tapped AC wall transformers and bipolar DC wall transformers aren't quite as easy to find.
Your use of two (or more) charge pumps running at the nominal 20-30khz may create issues such as heterodyning with each other....and or with the CE-1 clock (multivibrator) circuit.
To alleviate the charge pump/charge pump issue...you must sync the two charge pumps but....this will not solve potential issues with the CE-1 clock.
Look to the data sheet for instructions.
If you haven't read this thread....doing so might explain a few things.
http://www.diystompboxes.com/smfforum/index.php?topic=98225.0
I suggested the LT3467 because it seemed like it would work well for this circuit. Last year I worked up a modified CE-1 build as well and...and I plan to use this power supply.
Quote from: njkmonty on January 12, 2018, 06:52:07 AM
Yes that's how I measured. It. With my fluke dmm
What if you measure the current draw at the very first 9V input? It should be pretty much the sum of the two rails + some extra draw coming from the components in between...
Anyway, since you etched your build based on an original layout I find it hard to believe that something is going wrong, especially because the effect is working as expected.
Plus, since with your tests the LT1054 is performing well without noise issues, I don't really see the point of searching other stuff to power up the circuit. It works, it's compact and "builder-friendly". I guess that other solutions that include inductors or not so well famous to the diy community ICs wouldn't be the best way if you want to sell it as a kit. Plus, I guess that the double LT1054 is the most compact solution, excluding SMD stuff. The thing that I still find it hard to believe is how it is possible that two "parallel" LT1054 don't work while they do in series. It's really a mystery.
_________
Quote from: njkmonty on January 12, 2018, 12:07:34 AM
the following schematic i used got both rails on +/- 14 under load and seem to work fine,
i didnt hear any noise when playing the chorus, and appeared just the same as my 2 x lt1054 in series layout that worked for me,
i guess this allows each surge pump to handle the amps better?
also since the chips max can handle is 100ma, then i can use a smaller regulator T0-92 package, which is also 100ma rated.
i measured the Dc amps being used on circuit straight out of the (9V Boss power supply and this chorus seems to draw about 170ma?
(https://s5.postimg.org/4j2ifketj/Screen_Hunter_06_Jan._12_14.57.jpg)
I am sorry, I didn't read your edit. This layout seems much more reasonable to me, you can have at least 100mA per rail which is a really good achievement. But, what about onboard space?
170mA is because you are using 3 pumps to get your desired voltage... LT1054 have their own draw
I took the liberty to contact someone who is very familiar with the CE-1 circuit....Tim Larwill of Retro-sonic.
I asked him if he recalled the current draw of the positive and negative rails of the original CE-1.
This is his reply:
"I designed for a total max of 200ma, but 50ma per rail is enough."
I understand that reply to mean...the draw is below 50ma per rail.
I still believe that there might be a problem with the build.
Just because it appears to be working correctly doesn't mean that there isn't a possible problem causing excessive current draw.
This might be the reason why the voltage sags with a single charge pump power supply.
There could be a cold solder joint somewhere...or a solder bridge.
Are you still working with the board that you forgot to flip....and treated like a surface mount build?
here is some of my voltages
ive looked over pcb
everything sounds good, lfo and all leds seem to work fine including clipping led. i checked to see if any electrolytic caps were shorting, but no?
voltages due to lfo obviously vary, the setting when measuring was on vibrato fastest so i could actually get a reading
MN3002
1 +4.5
2 -5.1 --- -6.0
3 -3.4 --- -3.5
4 -13.25
5 -
6 -
7 -
8 -
9 -
10 -
11 -14.2
12 -6.0 --- -6.3
13 -4.1 --- -4.4
14 -4.1 --- -4.4
1458 (lfo)
1 varies -12 --- -13
2 -
3 varies -4.8 --- -5.8
4 -14.21
5 -
6 -
7 varies -0.8 --- +0.8
8 +14.27
ic 1 Ta7135p
1 +12.21
2 -
3 -
4 -12.99
5 -11.6
6 0.1
7 -12.78
ic 3 Ta7136p
1 +12.04
2 +12.04
3 +12.04
4 -14.2
5 -12.87
6 +12.75
7 +14.27
Shouldn't IC1 pin 7 voltage be +13V?
Shouldn't MN3002 pin 1 be +5V, pin 4 be around -13V?
How is it possible to have around +12V on IC3 pin 2 and pin 3?
thanks just found a solder bridge ic3 pins 1 and 2
updated voltages
ic 3 Ta7136p
1 +12.04 now... + 13.70
2 +12.04 - 0.024
3 +12.04 - 0.024
4 -14.2 - 14.22
5 -12.87 - 12.90
6 +12.75 + 0.095
7 +14.27 +14.27
ic 1 Ta7135p
1 +12.21 now... +12.23
2 - -0.003
3 - -0.005
4 -12.99 -13.0
5 -11.6 -11.65
6 0.1 + 0.143
7 -12.78 + 12.79
MN3002
1 +4.5 now... +4.5
2 -5.1 --- -6.0 -5.1 --- -6.0
3 -3.4 --- -3.5 -3.4 --- -3.5
4 -13.25 -13.25
5 -
6 -
7 -
8 -
9 -
10 -
11 -14.2 -14.2
12 -6.0 --- -6.3 -6.0 --- -6.3
13 -4.1 --- -4.4 -4.1 --- -4.4
14 -4.1 --- -4.4 -4.1 --- -4.4
Now you can try again simpler power supply! :)
but i measure the amps draw and it was still about 10-15 milliamps on positive rail
and about 60 milliamps on negative rail???
The negative rail will definitely draw more as it has both the BBD and Clock circuit attached to it but 60mA does seem high, when you tried the MN3007 did you remember to change the clock caps back afterwards? The clock will draw more current the higher the frequency its set to.
What kind of LED are you using for the peak detector circuit? An ultra-bright is going to draw more than a diffused red.
yes i replaced the correct clock caps
i also tested the current with the leds disconnected,
and it made little difference.
i plugged it in again, and the chorus sounded great????
if you have socketed all the active components or most of them, you could try measuring how much each of them absorbes by plugging one of them at the time. Once you find the one that absorbes too much we can start to troubleshoot that area because there still must be some sort of issue. Anyway, with the new lower readings, simpler charge pumps may start working (?)
the current draw hasn't changed
Try to unplug some of the active components while measuring and see which one makes the draw be so high
Any news?
The news is, we're both building another board each just to check everything and to compare etc. I've just about finished mine so expect an update soon.
Which transistor subs are you guys using/planning to use?
2SK30 -> ?
2SA733Q -> ?
2SC900F -> ?
2SA493 -> ?
2SC536F -> ?
Dunno yet. I'm going to try the socket and see approach!! :icon_biggrin:
Hey guys, while you are still troubleshooting the issue I went on by myself and draw a schematic & board layout for a 1590BB CE-1.
I have used a DC/DC converter that takes from 9 to 18V and gives +-15V. It has 100mA output per rail. The thing is that on its own it gives me some nice +14.8 and -14.8 but as soon as I hook it up to the circuit something takes the input voltage down around 11V and it starts to smell bad..
The effect doesn't work, I only have clean signal in every configuration. The peak level led and the speed led don't light up.. I am assuming that something is drawing too much around the 1458 area since it seems that it isn't doing its job. Could you please help me to troubleshoot this if I post maybe the schematic and PCB layout?
(Just to be clear, I am not getting any output from the MN3002 by probe testing)
do you have a real mn3002? or one of those da converter eBay things?
What's these da converter eBay things you mention?
MN3002 Package:DIP-14,High-Accuracy 8-Bit D/A Converters
Quote from: njkmonty on February 04, 2018, 08:25:15 PM
MN3002 Package:DIP-14,High-Accuracy 8-Bit D/A Converters
I know what you mean, I have the real ones... I got those from a guy who works for Malekko, so they should be good. I could still try to sub one into my original CE-1, but I am afraid of socketing the original MN3002 inside of it.
Anyway, I measured the draw and it seems like 65mA on the negative rail, and 10-20 on the positive. Looks like the 1458 isn't generating the LFO though, otherwise the modulation LED would blink no matter the MN3002
Oh right, never knew of these things, same part number as the bbd.
These..
https://rover.ebay.com/rover/0/0/0?mpre=https%3A%2F%2Fwww.ebay.co.uk%2Fulk%2Fitm%2F273047649965
Quote from: DFX-PCBS on February 04, 2018, 08:32:54 PM
Oh right, never knew of these things, same part number as the bbd.
These..
https://rover.ebay.com/rover/0/0/0?mpre=https%3A%2F%2Fwww.ebay.co.uk%2Fulk%2Fitm%2F273047649965
Yes I have those even if I didn't get them through eBay. They seem legit. But there is something strange around the circuit... I have tried as well with the LT1054 charge pump on a breadboard and I have the same results. It seems to draw too much no matter what I do. I still don't get why the 1458 isn't generating the LFO wave..
my version draws that amount and sounds great
What's your Z impedance seen if you hook up your DMM on +14 and -14 (without supplying it) seen by the meter?
And what about +14 to 0 and -14 to 0?
I have
-14 - 0 = 8.90K
+14 - 0 = 15.13K
+14 -14 = 6.73K
I was thinking... are there any resistors which need to have a strict wattage? Like the ones linked directly to 14V... could a lower wattage resistor cause draw issues and voltage drop to my DC/DC?
Okay guys, I've troubleshooted my build and now it powers up nicely with no issues. It draws aprox -25mA on the negative rail now that I fixed it.
Although I still have two issues:
no peak & rate LEDs light up
Vibrato is amazing, but switching to chorus provides no effect.
Any suggestions on what to look at?
check led voltages and try different orientation aswell as q14
check around ic4a measure volatge at d5/r35 are their any voltages when switching to chorus? if not trace back through switch to ic 4
Okay guys, made it work. Thank you for your support.
In the end it only draws maximum -30/35mA and +15/20mA. I have tested the LT1054 pump and it powers it up without issues, although I prefer using a DC converter which saves much more space and it's much more reliable (even if it costs about 15€).
One thing that I don't get it's the trimpot... what does that adjust?
Quote from: willwaush on February 08, 2018, 05:58:47 PM
One thing that I don't get it's the trimpot... what does that adjust?
The balance between the two outputs of the BBD.
It is also called a clock null trimpot for cancelling the clock signal noise.
This can be adjusted by ear but...for best results...an oscilloscope should be used.
With a scope connected to the trimmer's output...while adjusting the trimmer...you can visually see one BBD output superimpose over the other.
Great, well done Willaush!! Glad you got it going. I nearly have a working board too and was wondering about the trim pot. Havent touched it yet and I dont have an Anuscope .
By the way, this is how I got my power section working:
---> Variable Voltage Regulator (LM317) ----> +14v
/
18v Wallwart
\
---> LT1054 -----> -14v
When biasing both the LT1054 and the LM317, after doing the maths, I used trim pots in place of the biasing resistors. Under load, both rails dropped a little bit so I was able to re-adjust. I know this thread is entitled '9V to +-14v' and I've used an 18v supply. This is just a temp measure until I am satisfied my board is 100%
Quote from: nickbungus on February 09, 2018, 08:01:21 AM
Great, well done Willaush!! Glad you got it going. I nearly have a working board too and was wondering about the trim pot. Havent touched it yet and I dont have an Anuscope .
By the way, this is how I got my power section working:
---> Variable Voltage Regulator (LM317) ----> +14v
/
18v Wallwart
\
---> LT1054 -----> -14v
When biasing both the LT1054 and the LM317, after doing the maths, I used trim pots in place of the biasing resistors. Under load, both rails dropped a little bit so I was able to re-adjust. I know this thread is entitled '9V to +-14v' and I've used an 18v supply. This is just a temp measure until I am satisfied my board is 100%
That's not an issue since you can always double your 9V to 18V when you get your board working! I simply used diodes instead since, like you say, the voltage drops down a little bit on his own anyway.. so with usually 2-3 diodes you get exactly the +-14V.
I don't have an oscilloscope as well, I need to figure out how to set it by ear.. I am noticing a tiny little volume drop when I engage the effect, I am wondering if it's caused from this trimpot (?). anyway, it's barely hearable as drop, not a big deal. I was thinking as well of lowering R46 and R90 to get more signal pass through, but I'm not sure if the tone of the chorus remains the same.
Anyway, it's really clear how if you design a good board with the right links you can use any kind of supply, even a double LT1054 isn't an issue at all. You guys should go on with it... it's clear how that etched board is probably having some issue that makes it draw more than expected
Quote from: willwaush on February 06, 2018, 09:01:13 PM
Okay guys, I've troubleshooted my build and now it powers up nicely with no issues. It draws aprox -25mA on the negative rail now that I fixed it.
Although I still have two issues:
no peak & rate LEDs light up
Vibrato is amazing, but switching to chorus provides no effect.
Any suggestions on what to look at?
What solved your issue?
I had an extra link onto the PCB design so I simply needed to remove that.
Leds didn't work because they were wired reverse
Chorus didn't work because since mine is a double PCB design looks like the pin didn't bring the chorus stage voltage to the MN3002... I had to run it with an extra wire to where it needed to go
Anyway, they were all issues related to my board design, nothing special!
Hey guys, does any of you know why on the "stereo" channel I have plenty of wet volume but on the "mono", where it blends automatically wet and dry, when I engage the effect I get a slight volume drop ? What should I look at?
Sorry, I don't but I was having an issue where my output 1 (the effect out) sounded horrible muddy and bassy. I had 2n5457s for the fets. I swapped one of them out for a j113 and it instantly sounded better.
I'm going to try the other 2 tomorrow to see if that makes a difference, but I remember buying some J201s and 2n5457s from China ebay when I first started and I've never been convinced they aren't dodgey.
Just another note on the powering of the CE-1. Heres the schematic (http://i999.photobucket.com/albums/af111/njkmonty/ScreenHunter_07%20Apr.%2001%2012.27_zpsxy2tvexj.jpg)
With my test rig as it is with pots controlling the output voltage on both rails:
---> Variable Voltage Regulator (LM317) ----> +14v
/
18v Wallwart
\
---> LT1054 -----> -14v
I can control the voltage on both rails. By playing with the pots, I can really fine tune the lfo to give me the most swing.
This has led me to question whether I have a mistake on my board as there are no trimpots or other methods for fine tuning the lfo on the schematic.
Could it be so sensitive or if I have made a mistake where would be the place to look.
p.s. Could subbing the transistors make a difference. I cant remember what I used as I'm at work at the moment but I will post them tonight with my voltages.
If you are referring to changing the circuit voltage with pots....
I believe that I mentioned early on that deviating from the stock CE-1 voltage of +14/0/-14 would change the LFO and the clock from the original.
It sounds like that is what you are encountering when you adjust the voltage.
More width may sound good in the vibrato setting...but not so much in the chorus.
If it sounds good to you....then it is good!
I've been messing with this again recently and I've been having great results with the MC34063 (https://www.sparkfun.com/datasheets/IC/MC34063A.pdf). The results have been much better than the LT1024. (Less drop, ripple and noise)
Is there anyway I can get both +14 and -14v out of one MC34063? The inductors are relatively large and having one less chip would free up some room.
Quotes there anyway I can get both +14 and -14v out of one MC34063?
There's a couple of applications notes mentioned at the top of the datasheet.
Here's one. You can see a +/- supply example. The + is regulated by the switcher and the - is made-up then regulated linearly.
http://www.intusoft.com/onsemipdfs/AN920-D.pdf
Have a poke around at other switch-mode chips. There's many tricks to get a + and - rail.
I think some of the Linear-tech devices show some more examples.
QuoteThe inductors are relatively large and having one less chip would free up some room.
The switch frequency looks quite low by todays standards (200kHz pretty common).
The low frequency pushes the inductor size up.
Thanks Rob. I've had a quick look at the link and I dont currently (or voltagely !! :icon_redface:) understand, I'll have a good read tonight in bed and see if I can expand my knowledge.
QuoteThanks Rob. I've had a quick look at the link and I dont currently (or voltagely !! :icon_redface:) understand, I'll have a good read tonight in bed and see if I can expand my knowledge.
The circuit I was looking at was fig 31. I guess it's more of an example that it can be done. That circuit is still a bit messy. I know some of the linear technology applications notes and datasheets have other often cleaner examples.
The high-end Yamaha basses used ICL7662/ICL7660 with two inductors to get +/- rails. They are quite low frequency, so require largish inductor values. People did complain of the 10kHz whine - so not recommended.
Shielded inductors are a good investment to prevent noise problems.