Max1044 - You are the bane of my existance

[chromesphere]

Hey everyone,

I'm building a few rangemasters on my own layout and i've used the max1044 to supply -9v to the circuit.  I'm finding after boxing them up, 2 of 4 have dead max1044.  I get the smell of burning electronics.  I touch the max and its red hot.  Test the max, its dead. A long time ago i built an FSH1 with a max1044, it did the exact same thing.  Worked for a couple of hours and then died.  I used an LT1054 instead and its still going fine 3 years later...

The question?   Is the max1044 a poorly designed IC?  Is it poorly suited to 9v stompboxes (or even just the rangemaster and the FSH)?  I heard that they should be ok in low current drawing circuits (which i assumed the rangemaster was), but maybe its drawing too much current? Do you have problems with the max too, or am i just unlucky with it?  The only reason i bought them was cause tayda dont have the LT1054...i should have just bought the LT1053 from somewhere else, im positive the max is the problem yet again...Just seems all the stuff i build with LT1054 keeps working forever, but max builds dont last long...


Paul

[Govmnt_Lacky]

Hey Paul,

I would be quick to point out the current limitations on the MAX1044 charge pump but, I would doubt that the Rangemaster circuit draws more than 10-15mA  Also, I would need to point to the PT-80 circuit (GGG) and the fact that it uses the MAX chip and that it more than likely draws more current than the Rangemaster. 

Since you are using your own layout, could it be a design flaw in the way you are implementing the MAX chip? Is your charge pump portion of the circuit identical to other "problem" designs? Bounce it off of the MAX datasheet and see if you can find differences.

Good Luck 

[chromesphere]

Hey, thanks for the response!

I actually just edited my post to say "surely current draw couldnt be an issue".  I've checked over my layout a million and five times and it checks out...it's crazy, whenever i touch a max, it dies...Maybe i shorted something out (twice) inside the enclosure....i think their just too dam sensitive maybe the LT1054 can just take a bit of man handling...idk...

Paul

Edit: im going to sleep now to dream of having a punch on with a giant IC called max. Five hours of fault diagnosing does it everytime

[kwijibo]

Have you only seen this with units from Tayda?

[R.G.]

The MAX1044 is *almost* an incredibly good part, with one big problem in today's pedal market. Your problem is almost certainly the fact that the 1044 has an absolute maximum input voltage of 10.000000V Above that, they die. The LT1054 has a higher current rating, which isn't needed for a rangemaster, but also a higher tolerance for input voltages. The average guitarist of today has been "educated" by net-blather that if you take pedals designed to run on 9V batteries and power them with 12V, 15V, 50V, whatever, it will magically make their fingers play guitar better. 

[Toney]

 That much heat means dead short amounts of current inside the MAX or out. Please show an example of how you have it laid out.
Maybe that thing you're sure you checked 100 times needs another set of eyes?
I have had so many "doh" moments.  

[iccaros]

to add to R.G.
I was having MAX 1044 die allot also, problem was over voltage, not that I was trying to get higher voltage, just a lot of designs on this board showed them connected at 12V
but I changed over to the intersil 7660S

there are about $1 each in small quantities. and the S version handles higher voltage. I have not had one die yet... Did I just say that.. where is some wood...

[Colonel Angus]

Thanks RG! That explains why my crappy Wallwart fried the MAX in my Gristleizer...

[R.G.]

This kind of thing is one reason I habitually roll my eyes at the rising storm of internet guitarists (!?) that think if it's designed for 9V, it will sound really cool on 18V or 24V.

I have gone to defensive designs in our pedals, assuming the worst about the owner's behavior when they get the pedal. I presume the new owner will immediately stuff any voltage they can get on a connector into the inputs, outputs, power jacks, etc. It increases the part count, but then reliability is one of the things I work hard at.

[Colonel Angus]

LOL @ R.G "internet guitarists", so true, there are ups and downs to the forum lifestyle. Urban myths become populated on a much larger scale. I can assure that my wallwart was marked as 9V, I wasn't trying to get cute and try 18V. Just a scuzzy wart

[R.G.]

oops... sorry, that wasn't directed at you. 

We do get a lot of people calling up to ask if our pedals work on 12, 18, 24, etc., but with the MAX1044, even a little voltage over 10.0 kills it, so a poorly regulated 9V wall wart will kill them, them all right.

[Bill Mountain]

Rick from FCUK has ran a Maxx at 12 volts in some of his designs with no problems.

I just wanted to point that out.

I have heard that Tayda's parts are a little suspicious.

[R.G.]

Rick from FCUK has ran a Maxx at 12 volts in some of his designs with no problems.

I just wanted to point that out.

Semiconductor batches vary. I've seen batches of 1044s work at 12, and also some that fail at 10.01. 

Pushing semiconductors beyond the absolute max on the datasheet is always chancy. Often it works. The times it doesn't are what get you, infrequent as they may be.

[kodiakklub]

i say let those idiots blow up their pedals. if it says 9VDC on the pedal, guess what im going to plug into it. if the microwave says 120VAC guess what im going to plug into it. i wonder if those internet guitarists also go to home depot to buy an electric stove and ask if it can run on 120VAC. Oy.

[.Mike]

I have heard that Tayda's parts are a little suspicious.

Ugh. This is how unsubstantiated rumors get started and perpetuated. I have not had a single problem with any of my numerous orders from Tayda. They guarantee everything they sell:

Buy with confidence. All our products comes with 100% satisfaction guarantee, any damaged or malfunctioning product will be fully refunded including shipping costs without any hassle.

I challenge anyone to find a US supplier who has put a similar policy in writing. I haven't seen one yet. Some don't even guarantee that you will ever receive your order!

Until Paul indicates otherwise, the first and most obvious suspicion is user error. All of my unregulated 9v adapters come in over 12v, and most are well above that. That would probably fry a 1044 no matter where you bought it.

Mike

[iccaros]

this is why I suggest the  7660S if you plan on using more than 9v but not more than 12v.  It will fit in the same design as a MAX1044 and even has the booster ,unlike other options.

too bad I do not know of a way of ensuring the chip only gets 12v max, so if someone plugs in a 24v it will put out 12v and if they put in 12v it will put out 12v.  There is a way, I just don't know it

[R.G.]

too bad I do not know of a way of ensuring the chip only gets 12v max, so if someone plugs in a 24v it will put out 12v and if they put in 12v it will put out 12v.  There is a way, I just don't know it

Shunt regulator. You can come very close with an LDO.

[chromesphere]

Im back from my sleep and indeed dreamt about the max1044, except it had a moustache and we were engaged in a mexican stand off.

I dont think tayda units are the problem.  I actually have 3 max's.  2 are dead.  One of the blown units was from a previous batch (from futurlec probably) the other is from tayda.  I can tell because the tayda max's have a line in the markings on them (probably revised over the last couple of years) where as the futurlec max's (from 2+ years ago) dont.  One of the tayda max's still works. "yay".

I was thinking about some of the other pedals i've built with the max1044 in the past.  A Klon clone, which i never could get working in the past (may not be the max in this case though).  A boneyard with an LT, which is rock solid.  flanger 301 LT -> rock solid.  Maybe my power supply spikes up over 10v's or something and then the max, kind of locks up and just continuely heats up...?  I had a MB bumblebee (baldwin burns buzzaround) with another charge pump infront of it when i first hooked the RM's up.

I think i will just replaced them with LT's and call it quits, (or the 7660s) and never touch a max again.

Heres the layout that i made.  Its not the best layout you'll see (pot's in the middle of the board for example) but yeah, i dont see anything 'fundamentally' wrong here...i've gone over it soooooo many times...sorry for the bad layout i'm still learning diptrace.




U1 = PNP GE
U2 = anything thats not a MAX1044
C7, C5 = 10uf   
C4, C3 = 47uf
C9 = 100uf
D1, D2 = 1n4001
the rest is range master in out caps, bias resistors etc

[chromesphere]

I think what im trying to say here is, i've never had an LT (or TC) die on me...ever...I think the MAX is just too fragile for diystompboxing...well....atleast they are when im using them

[Earthscum]

After reading this throughout the day, I'm wondering if a 5V regulator could be used as over-voltage protection. It can definitely handle the "typical" overvoltage from a "power supply boost", plus some. Questions would be: will it play nice with the MAX, and will an "L" be fine (less than 100mA), or does the MAX require  a 1A.