PEAK Atlas DCA75 Pro

Thread Starter

SamR

Joined Mar 19, 2019
5,031
Been looking at transistor curve tracers for a while and came across this DCA75. A bit pricey @ ~$170 but a heck of a lot cheaper than some of the alternatives and seems to do much more including connecting to the computer to do graphing and analysis. Does anyone have any hands-on experience with one?
 

dl324

Joined Mar 30, 2015
16,839
Needing to be connected to a computer to display the curves would be a showstopper for me.

I bought a Jud A Williams curve tracer on eBay recently for under $50; though most curve tracers are being priced in the $200-600 range. This is for used equipment, sometimes in unknown working condition.
 

rsjsouza

Joined Apr 21, 2014
383
I have been looking at the DCA75 Pro for a few years now and I almost pulled the trigger many times. It really seems a well done device and the interface with the computer is a great way to document and compare different devices, share data, save settings, etc.

However, the major drawback to me was always the maximum useable voltage - something around 12V if I recall correctly. To me this is very insufficient to get some decent curves for higher power transistors and higher voltage zeners, for example.

In the end, I got this last Christmas an off brand curve tracer that can go up to 36V @ 2.5A, which is not ideal but much more suited for what I do. This curve tracer was designed many years ago by a Chinese developer named locky_z and it works really well - the software has minor translation issues here and there but it is rock solid, quite customizable for the DUT and fully featured. It also has a useful component identifier as well.

A few references:
A bit of history and links: https://www.eevblog.com/forum/testg...nd-optimization-of-locky_z-transistor-tracer/
The github project with all HW blueprints and manuals: https://github.com/locky-z
His shop (ships from China): https://www.ebay.com/itm/Locky-z-2019-Curve-tracer-with-new-component/114234796434

I really have no relationship with the guy, but I really liked his product.
 

dl324

Joined Mar 30, 2015
16,839
There was a person who designed a DIY curve tracer and made the schematic available. Can't remember his name. He was also involved with the toner transfer for making PCBs. I was tempted to build one, but I got a broken BK Precision 501A on eBay before I got around to it. Even though it's a very useful tool, there's not really much to it.
 

sparky 1

Joined Nov 3, 2018
756
The peak atlas pro dca75 specs are impressive. This small device test parameters are in the range of ±12 V, ±12 mA.
When you are shooting for precision the software makes it easier to get closer tolerance than not having curve trace software.
If three different testers all give different results which one will you trust.
The higher voltage of the Locky z, sys comp circuit gear for tubes, utracer, and others.

Not to forget about the diode. The basic IV curve has a point near the knee of the curve that is important mathematically.
In setting up a photovoltaic system the performance can be analyzed using IV curve. This is being taught in better schools.
 
Last edited:

Thread Starter

SamR

Joined Mar 19, 2019
5,031
I've looked at Octopus rigs and even used a sig-gen with a scope for the DUT but it seems they only do one trace at a time. Which would have to be transposed to a graph trace by trace. I would like something a bit better. The DCA75 does a whole lot more than just curve trace tho...
 

rsjsouza

Joined Apr 21, 2014
383
The higher voltage of the Locky z, sys comp circuit gear for tubes, utracer, and others.
36V is not nearly enough to drive the anode or G2 of vaccuum tubes. locky_z's range is well within the work region of common power transistors such as TIP31 or BD135, which have a breakdown specified at 40 or 45V. If you don't need that, then the DCA75 is for you - it is just one aspect that I don't see mentioned very often when people are trying to decide on a curve tracer.
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
No, they don't and I hadn't even considered the max test voltage. It is something to take into consideration before plunking down $150-200.

EDIT: Considering what I am currently doing (low voltage experimentation) I'm going to take a flyer on it and order it. Looked at several Utube reviews and shootouts of various curve tracers plus my experience in doing some "Octopus" testing using a sig-gen and scope the DCA75 seems to have the advantage albeit not without cost. YMMV
 
Last edited:

Thread Starter

SamR

Joined Mar 19, 2019
5,031
It shipped and should be here Monday. This is their "official" UTube demo which pretty well covers it. I'll add any other things after I actually use it. It runs off of a single AA battery.
 

sparky 1

Joined Nov 3, 2018
756
Sam, you can cut and paste the data into Excel spreadsheet and extend the I/V graph to what ever voltage and current you want.
The ability to obtain a precision digital data file on an actual component is what allows a wide variety of analysis in semiconductors.
In LTspice and other simulation software are parameters for a transistor's model. While most parameters are known for a specific
transistor there can be variations in the actual part and the atlas will give you specific data close to 25 degrees C
You can also build a simple test jig for matching 2 transistors, I recall using plastic tweezers for some tests.
I have an old thermal scanner for shooting the temperature at a pin point, the test jig the pcb ect.
If the scanner reads 24.8 and it agrees with the datasheet thats nice but I use it very often. I let the part cool on a surface then as the temperature
approaches 25 i push the test button. It is just the nature of semiconductors. The price on thermal scanners is dropping because pandemic.

 
Last edited:

Thread Starter

SamR

Joined Mar 19, 2019
5,031
Yup but the device itself limits the Voltage itself to 12V max which it generates from a 1 1/2V battery. If you wish to compare several devices then the cut and paste to Xcel of each device is handy to build a chart comparing them. I just hope it is as handy as the ads lead me to beleive. I'm a "show me" kind of guy.
 

sparky 1

Joined Nov 3, 2018
756
After sweeping the transistor then back on the solderless breadboard you have your supply connected
so the voltage is constant. On the base you could use a a precision voltage-controlled current source.

(Thomas Floyd used battery symbol and a resistor explaining transistor's base current) differs some from the larger current variety, not too precise.
http://electronicsbyexamples.blogspot.com/2013/05/adjustable-constant-current-source-1a12v.html

What does a precision CS look like (which is abbreviated in textbooks with different symbols controlled current source):
Figure 2-30 shows a a precision voltage-controlled current source. Floyds curve ball is still valid for it's math, the small battery might look like this.
https://www.sciencedirect.com/topics/engineering/controlled-current-source
 
Last edited:

Ian0

Joined Aug 7, 2020
9,667
I just bought one, and it is rather impressive. If they made a bigger one that would test power transistors at a decent amount of current, I'd buy that as well. (And then a third one if it could test valves)
The challenge is to connect it to a SOT23 transistor. It's just about possible to touch the probes on a SOD-123 diode and press the start button with your little finger, but for surface-mount transistors you need either a test jig or an accomplice!
Now here's a thought - could someone write a piece of software that turns the output into a SPICE model?
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
DCA75 came in today and playing with it a bit. Still reading the manual but here are 2 2N3905 transistors compared. Surprising difference (to me at least) in gain. Just a random sample from stores. The device itself isn't much to look at but what it sends to the software on the computer and what it does graphing it is impressive to me. Default is 51 samples per line.
1615269892289.png1615269619167.png1615270015082.png1615270059084.png1615270118959.png1615270164068.png1615270233401.png1615270289790.png
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
One other neat thing about the graphs it produces is that by placing the cursor over any point on the trace lines it gives the values at that exact point. Sad that it is limited to 12V max. Surprisingly it will also analyze TO-220 case style regulators.
 

sparky 1

Joined Nov 3, 2018
756
@Ian0, To solve for HV transistors can be found using Algebra to find the Y intercept.
All the curves have slightly greater and different slope with ordered pairs along the flat part of each curve.
Because all the curves have the same Y intercept so that any point can be derived.
Solve for Slope & Y-Intercept Y=mx+b and m=(y2-y1/x2-x1)
 
Last edited:
Top