Cheaper, (possibly) better MAX1044 alternative with boost

[.Mike]

Hi folks,

In doing some research for my upcoming builds, I stumbled across an alternative to the MAX1044. It's sold by Mouser, although it is not listed in their catalog. It is the Microchip TC1044SCPA or the TC1044SEPA: Mouser link, datasheet [PDF].

What sets this apart from other alternatives is that it has the boost feature that most of the other alternatives lack. It can also handle a higher voltage operating supply than a MAX1044, and it is about half the cost.

	Maxim IC MAX1044	Microchip TC1044SCPA
Supply Voltage	1.5v - 10v, 10.5v max	1.5v - 12v, 13 max
Frequency Boost	Yes, 6x	Yes, 45khz
Current Output	Same*	Same*
Price	Mouser: n/a Newark: 1@$2.30, 25@$1.82 Digikey: 1@$2.68, 25@$2.02	Mouser: 1@$1.12, 10@$.093 (TC1044SCPA) Newark: 1@$1.60, 25@$1.60 (TC1044SCPA) (direct ship from UK, $20 ordering fee) Digikey: 1@$1.28, 25@$0.78 (TC1044SEPA)

*Best Rick (frequencycentral) could tell. I'm still fairly clueless.

I haven't tried these, but I do plan on ordering some. I'll report back, but it might be a few months. In the mean time, if anyone else tries them, please let me know how they work!

I hope that helps someone.

Mike

[R.G.]

Go look up the Linear Technology LT1054.

[.Mike]

Thanks, RG. I've actually read about that before, and have read the datasheet. I am a beginner though, and as such, I'm still fairly overwhelmed when reading and understanding datasheets and how these circuits work. I'm learning, though.

Since there are already a ton of designs here with the 1044, I think sticking with something that is relatively proven is probably a good idea, at least for a beginner like me. I would love to try the LT1054, but not until I have a better understanding of what I am doing and why I am doing it. I know they're pin compatible and that the LT1054 can provide more current, but the TC1044S is still about half the cost of the LT1054.

Give me a year of learning and building under my belt, and I'll gladly try out the LT1054... heh! For now, a cheaper alternative to the MAX1044 is a good thing.

Mike

[Andi]

I've been using the ICL7660S. Which works. I'm not sure I'm using the right/best diodes though - is there a ebst type for these applications?

[frequencycentral]

@ R.G. - A lot of folks aound here get a lot of info from GEOFEX, me included. Respect!

I'm using MAX1044 because I came across the GEOFEX article - though there are now better alternatives. I assume things have moved on since that article was written - maybe time to update that article to include stuff like the Linear Technology LT1054 and Microchip TC1044?

I'm not having a pop - just pointing out that your word can launch a thousand ships!

[big bustle]

any word on this cheaper alternative? how has it worked out?

[frequencycentral]

any word on this cheaper alternative? how has it worked out?

I didn't use any yet!

You could try 555's: http://www.diystompboxes.com/smfforum/index.php?topic=80007.0

[Skruffyhound]

I was hunting around for a replacement at RapidOnline in the UK, and I came across the  ICL7660SCPA CMOS voltage converter. Comparison of datasheets showed the same functions as the MAX1044 but that it was good for 12v, it has the freq. boost and same pinout etc.
Maybe it's the same thing again, but I paid 5 pounds 80p for ten of them which is a fraction of the cost of the 1044 and Rapid delivers free in the UK over about 20 quid (AFAIR). Unfortunately I live in Denmark so I still get to pay , but anyway!

Sorry I haven't tested yet either but I have a bipolar supply (with a 1044) in the breakout box for my breadboard rig, I'll try and get around to switching them out.

Rapid Product code is 82-0788, go and read the datasheet there and see what you think.

[.Mike]

Do a search here on TC1044. It seems plenty of people have used them with no problems.

Mike

[tupinamba]

Hi !
So, Skruffyhound, did it work with the 7660SCPA ?... I'm planning on try it also but a feedback would be great...

[bluebunny]

I've done a couple of builds using the ICL7660SCPA and it seems to work just fine.  As I recall, the "S" is important if you want the frequency boost (so that's a "yes" for most of us here!).

[bluebunny]

BTW, @Skruffy (and anyone in the UK), the free postage at Rapid kicks in at £35 before VAT.  They usually get an order from me when I have plenty to buy!

[Dave W]

The problem I have is the output V of these charge pumps sags under load. Depending on what one is powering this can be "no big deal" or a "major issue".
Check the datasheets. Or just use your trusty DMM to see this.

[R.G.]

As they used to say TANSTAAFL - There Ain't No Such Thing As A Free Lunch.

Charge pumps do a remarkable job of making different voltages quietly with an IC and a few caps. But they do have limits, and the inability to source a lot of current is one of them. They - like all technological items - are not magic bullets. The devil is always in the details.

You have to know your application and the limits of what you're doing to address your needs. If you don't know how much current you need, even to an order-of-magnitude level, it's unlikely that you'll be happy with a random solution.

And you're right - always check the datasheet. That's kind of a generic comment for every electronic question.

[tupinamba]

Thanks for those answers...
What I would need is +9v / -9v for an AD781 chip. Do you think this would do the trick ?

[R.G.]

It depends on how much the AD781 rejects its power supply noise. There is some unavoidable noise on the output of a charge pump. You can get rid of this with a series regulator after it, but you lose some voltage. I don't know how much voltage the AD781 needs or how tolerant it is of noise on its power supply. The datasheet should provide some clues. Also how much current it uses. This will be critical. I'm assuming that if you're designing with the AD781 you're familiar with the datasheet, right?

[tupinamba]

Well, errrr 
I'm not a specialist and most of the time I follow pre-made layouts...

I don't know if this is what is important, but here is what the AD781 datasheet reads :

TMIN to TMAX, VCC = +12 V +/- 10%, VEE = –12 V +/- 10%
POWER SUPPLY CHARACTERISTICS
Operating Voltage Range ± 12
Supply Current 4
+PSRR (+12 V ± 10%) 80
–PSRR (–12 V ± 10%) 75
Power Consumption 95

As for the power supply noise rejection, there is a "Power Supply Rejection Ratio vs. Frequency", but I don't know if this is what it's about...

[PRR]

> there is a "Power Supply Rejection Ratio vs. Frequency", but I don't know if this is what it's about...

Yes. The pump runs at 6KHz or 45KHz. That's the crap which must be rejected.

64dB at 6KHz, 53dB at 45KHz.

For small signals, 64dB down from full supply at 6KHz is small but annoying.

53dB down seems worse, but the first-order leakage is inaudible. (Intermodulation may be an issue.)

The chip can draw 6mA. Wwe can probably sacrifice a part-Volt. 100 ohm 5uFd will give another 25dB rejection at 6KHz which should make "small" vanish.

Much depends on why you are using a S/H at all, and what you are doing with the output.

[R O Tiree]

As they used to say TANSTAAFL - There Ain't No Such Thing As A Free Lunch.

OT - I remember reading Robert Heinlein books when I was younger and came across that acronym several times and never knew what it meant (this was in the days before teh interwebz). Every day's a learning day...