THAT1646 SSM2142 Balanced Ouput Driver Comparison

Thanx Media.

The THAT part looks very good comparatively. How do these parts compare in price and availability? And just for grins, how do they compare at some lower freq like 25 KHz?

Peace - Out, Irv

Thanx for the report Media.

I was thinking of stocking up on some 2142's for future circuit mangeling, but I think you have changed my mind. Especially since we just clicked into November, I'll wait for the THAT guys for a little while. They don't have the best track record in town for hitting delivery dates.

Have you looked at or have an opinion of the ballanced to single ended input device? (INA134 or INA2134) Ive always just used an opamp for this but the component count and board realestate gains are beginning to look attractive.

Peace - Out, Irv

interesting

I haven't got any samples yet ... :sad:
as you may know ... I use ... err used to use many of the 2043 input and 2142 output chips and a drop in could interest me.

[quote author="mediatechnology"]Have you looked at the Bill Whitlock InGenius circuit for balanced in?[/quote]

I got some samples when they first became available. they measure very good and I thought they sounded good. I havent really done much with them though. does anyone know of a comercial product that is using these?

mike p

[quote author="Kev"]... use many of the 2043 input and 2142 output chips ...[/quote]
err
that should have been 2143
.. and 2141

While this series of chips have certainly simplified output design, I find their near universal use troubling, especially where long lines are used.

I've had no end of trouble with equipment using these chips in broadcast equipment in an RF environment especially during lightening events. The very low output impedance of these chips, with no output protection (either on chip, or in the equipment using it) do not like any kind of voltage pulse on it's output.

Although the facilty I'm working at now is very well grounded, when a nearby lightening strike hits (direct hits are handled through grounds; nothing gets destroyed), the voltage pulse in long lines between some studios and the rack rooms cause these guys to give up the ghost. To further complicate matters, only about 1 in 10 manufacturers use the dip version of this chip, allowing easy replacement; most are surface mount. This means I have to keep dozens of replacement assemblies (not just chips! - we're talking lotsa $$) in inventory and cringe everytime the wind whips up in summer.

They just don't build 'em like they used to..... :evil:

ok, rant off.

[quote author="mediatechnology"]Have you looked at the Bill Whitlock InGenius circuit for balanced in?

I haven't messed with them yet but he uses bootstrapping to get a high effective input resistance. This is to reduce common-mode errors from the receiver loading the source unevenly. I think it helps primarily at the low frequencies.

Both Jensen Transformers' site and THAT have info on it.[/quote]

No I hadn't. I have now. Thanx for the tip!! :guinness: :sam:

Sorry this response took so long. I've been off line for about aweek.

Peace - Out, Irv

there's a big announcement on the THAT website now:
It's out! The outsmarts balanced output driver ICs
http://www.thatcorp.com/1600desc.html

And another even more interesting newsflash:

"THAT Corporation has announced a partnership with Mouser Electronics that will permit better service to customers desiring smaller order sizes."

Ok, cool I didn't realise you had a shop, hey, how are those THAT transistor arrays compared to lm394 and MAT02?

Pretty soon THAT will have everything you need to build a pretty good quality mixer..

I haven't worked with the 300s on the bench, but from a datasheet review they do seem like a viable alternative to hand-matching discretes with pretty-decent noise performance.

Click to expand...

The DC precision is often (but not always) not a problem in audio, but the beta is pretty disappointing.

Samuel

[quote author="mediatechnology"]THAT are showing a 1K unit price of $2.25 in 8 pin DIP for the 1646. MOQ 50 pcs. ADI on the SSM2142PZ (RoHS) is $2.75 100-499. So they look to be very competitive price-wise. They have parts (obviously) but won't be shipping until late November I'm told because they're doing final qual.

I will shoot 20 kHz as soon as I get the chance; the 2142 even at 20 kHz is still a little rounded. With the 2142, the wave shape varies much more as one output is grounded or floated. Both the 100 kHz and 20 kHz waveshape, when viewed single-ended, improve with a ground-tied output in the 2142.

Although the 1646 changes waveshape a little bit as one end is switched from ground-tied to floating, its not as much and far more square. Haven't checked large signal performance.

The photos in the first post are with 600 ohm bridge-tied (tip to ring) with both ends floating.

THAT on their website have a comparison of the two parts where they show oscillation in a 2142 output driving ground in a typical balanced to unbalanced connection. http://www.thatcorp.com/datashts/OutSmartsComp.pdf

I shorted the 1646 to ground through a 22R and that ouput looked very well-behaved.

I remember some balanced outputs, the Otari MTR-10 comes immediately to mind, that would dump so much current into ground when unbalanced that they were fantastic cross-talk generators. The 1646 appears to overcome this quite well.[/quote]

If you are messing around with these on the bench an off-sheet parameter to look at is turn-on/off characteristic. Back when I worked at a fairly large company I give sundry IC companies my wish list at their annual dog and pony shows where they pretend to listen.

My wish list included.

1) quiet turn-on/off characteristic or possibly mute pin so you could externally control this.

2) modest forward voltage gain (perhaps 3 dB) and ability to run from hotter unregulated rails so you could squeak out a few extra dB of signal swing.

3) realistic pricing. To get designed into high volume apps they need to compete with a couple of opamps and precision resistors.

Properly dealing with the turn-on/off transients which could be done using conventional process technology in an IC might be enough to justify the part's cost premium. 100 kHz square waves was not on my list (more like 30 kHz sine wave power badwidth with no HF squirreliness).

Since I never saw a satisfactory response, I guess I was asking for too much at too low of a price point. I appreciate the cost to tool even a relatively simple IC like this, for what I suspect is a disappointing market volume. These don't exactly get designed into IPODs.

JR

That's good news about the distribution by Mouser.

My only quarrel with the THAT Q arrays is their relatively low beta. If they could make a multiple of a part like the Toshiba 2SA1316 or 2SC3329 I would be very happy indeed.

As an aside; there is also a balanced in/balanced out version of these chips (the 1606) which unfortunately comes in a QSOP package..... they are so small you can hardly see them !!

We will be producing a small adapter PCB which allows you to wire these up easily.... it will be supplied with the chip and surrounding components already soldered to it so that you can deal with it as though it were a DIL package device.....

colin
www.audiomaintenance.com

[quote author="Slenderchap"]there is also a balanced in/balanced out version of these chips (the 1606) [/quote]

cool. this could be of use to me. Im going to request samples. do you know what the CMR performance of the INPUT side is like?

Adapter PCB for the unbelieveably small 1606....

http://www.audiomaintenance.com/downloads/that1606_adapter.jpg

The dil pins will be supplied but not fitted....

[It has the components that are shown in the datasheet.... 3 x 100nF caps and 1M0 resistor... + 1 x THAT1606 of course].

Colin
www.audiomaintenance.com

[quote author="Slenderchap"]Adapter PCB for the unbelieveably small 1606....

http://www.audiomaintenance.com/downloads/that1606_adapter.jpg[/quote]
Do you offer these for sale ? I couldn't find them in your online store.

I was going to do a similar board (as I like to be able to swap out blown line driver/receivers, like Wayne says), but there's little point in duplicating work.

JDB.

JDB.

The 1606 adapter will be available for sale..... only a small percentage of the products appear in the on-line store...

just send me an email of your requirements

Colin
www.audiomaintenance.com

The 1606 adapter now appears in the on-line store...

part number AML-16-005

Colin
www.audiomaintenance.com