# A riddle involves designing a circuit

#### AnalogKid

Joined Aug 1, 2013
8,487
Some game show circuits.

ak

#### crutschow

Joined Mar 14, 2008
25,117
Why wouldn't they be expecting a detailed circuit? It's a simple problem that you should be able to solve in less than five minutes.............
It may be for you, but perhaps not to us mortals.

#### AnalogKid

Joined Aug 1, 2013
8,487
C'mon... you whipped this in a thread in January.

Speaking of...

A thread this calendar year regarding a game show circuit (but not necessarily by that name) where someone posted the world's simplest version. It used neon bulbs as both the latching elements and displays. I can't find it, but I'm rarely successful at finding past threads.

anyone...?

ak

#### WBahn

Joined Mar 31, 2012
25,913
It may be for you, but perhaps not to us mortals.
If we are at the point when it is unreasonable to expect an applicant for an EE job to be able to come up with something like

during an interview, then is it any wonder that companies feel the need to outsource engineering jobs to countries where this would be considered something that a sophomore taking a first course in digital logic should be able to do easily on ten-minute pop quiz.

#### crutschow

Joined Mar 14, 2008
25,117
Some game show circuits.

ak
In the second circuit with relays I believe there needs to be a diode in series with each PB so that power is removed as soon as the first relay closes, avoiding any voltage feedback through the closed PB.
For example, if S1 was pushed when S2 was still engaged after being initially pushed, the latch would transfer from the 2nd relay to the 1st.

#### crutschow

Joined Mar 14, 2008
25,117
If we are at the point when it is unreasonable to expect an applicant for an EE job to be able to come up with something like
........................
during an interview, then is it any wonder that companies feel the need to outsource engineering jobs to countries where this would be considered something that a sophomore taking a first course in digital logic should be able to do easily on ten-minute pop quiz.
I think you are being overly optimistic about the ability of most sophomore's ability, taking a first course in digital logic, native or foreign (although I'm sure you would have had no problem), but I think companies outsource to save money, not because native engineers don't have the ability to do the job.

#### AnalogKid

Joined Aug 1, 2013
8,487
In the second circuit with relays I believe there needs to be a diode in series with each PB so that power is removed as soon as the first relay closes, avoiding any voltage feedback through the closed PB.
For example, if S1 was pushed when S2 was still engaged after being initially pushed, the latch would transfer from the 2nd relay to the 1st.
Nope. Relay power comes through the three NC contacts in series. When any relay changes, power to the other coils is broken. Power to the selected relay comes through the NO contact. The circuit is dependent on the relay armature having enough inertia to move from NC to NO without continuous power to the coil. Not my circuit, but points for simplicity.

ak

#### MrAl

Joined Jun 17, 2014
7,665

Hi,

You need to add some kind of storage to your circuit so the circuit can detect the first horse only and also remember it.

Very often we need to know which one came in second and which one came in third also, so you might need to detect three different arrivals and store all three of them. So for three horses A, B and C, they might come in as either of:
A,B,C
A,C,B
B,A,C
B,C,A
C,A,B
C,B,A

For example B,C,A means horse B came in first, C came in second, and A came in third.
Although not as likely, there could be a tie too depending on the resolution of your timing.

Some people place a lot of money on these outcomes so if it doesnt work right they are going to get very mad at you

#### WBahn

Joined Mar 31, 2012
25,913
I think you are being overly optimistic about the ability of most sophomore's ability, taking a first course in digital logic, native or foreign (although I'm sure you would have had no problem), but I think companies outsource to save money, not because native engineers don't have the ability to do the job.
If expecting a person that has taken a first digital course to be able to design a simple circuit like this that uses an RS FF in a very simple way is being overly optimistic, then please describe the type of problem that such a person should reasonably be able solve when applying for a job.

Two (anecdotal) points for consideration.

2) Companies outsource for both reasons. The (small) company I worked for hired many H1B workers (bringing them here, so not true outsourcing by a strict definition, but effectively so) solely and entirely because we could seldom find a U.S. educated engineer (new grad or otherwise) that we were willing to hire -- and all we were looking for was someone that had sound foundational skills at about the junior level. Hiring an H1B worker is NOT cheap and you HAVE to pay them at least the market rate for U.S. workers and you then have to show that you are constantly trying to replace them with a U.S. citizen (or legal permanent resident) for as long as they work for you.

#### AlbertHall

Joined Jun 4, 2014
10,022
Nope. Relay power comes through the three NC contacts in series. When any relay changes, power to the other coils is broken. Power to the selected relay comes through the NO contact. The circuit is dependent on the relay armature having enough inertia to move from NC to NO without continuous power to the coil. Not my circuit, but points for simplicity.

ak
crutschow is right. Say button 2 is pressed, the relay switches but button 2 is still pressed. This feeds power from relay 2 coil to the button common, so if now button 1 is pressed then relay 1 will receive power via the two pressed buttons. Realy 1 will operate - oops!

#### AnalogKid

Joined Aug 1, 2013
8,487
Well..............look at that. Score 1 for Wally.

Puny defense: As I said, not my circuit.

ak

#### crutschow

Joined Mar 14, 2008
25,117
If expecting a person that has taken a first digital course to be able to design a simple circuit like this that uses an RS FF in a very simple way is being overly optimistic, then please describe the type of problem that such a person should reasonably be able solve when applying for a job.
I'm just saying that there are probably only a few people able to put such a circuit together with no experience in a few minutes while under the stress of an interview.
Of course, I understand that a person with your superior intellect would have no problem under those conditions.

#### WBahn

Joined Mar 31, 2012
25,913
I'm just saying that there are probably only a few people able to put such a circuit together with no experience in a few minutes while under the stress of an interview.
Of course, I understand that a person with your superior intellect would have no problem under those conditions.
So the question still stands -- what IS the hardest question that is reasonable to expect an applicant for an EE position to be able to solve under the conditions of an interview?

Would asking them to design a voltage divider that steps a 3.3 V signal down to 2.5 V (assuming a very high impedance load) and drawing approximately 1 mA from the 3.3 V source be unreasonable?

Apparently it is since fewer than about one-quarter of the new-grad applicants could do it.

#### KL7AJ

Joined Nov 4, 2008
2,225
Hi,

I'v been asked this riddle :
"Three horses are having a race. at the finish line of each lane there is a detector which gives a logical '1' when a horse crosses the line. design a circuit which will let you know which horse won the race"

I have a direction to solve this question but I'm not quite sure It's the best solution or the appropriate one so I would like to hear you suggestions.
My thought was to connect the 3 detectors (at the end of each lane) into the input of a decoder so by the output of the decoder we will know which horse won.

Thanks!
A simple XOR should do the trick

#### crutschow

Joined Mar 14, 2008
25,117
A simple XOR should do the trick
How would that work for three horses.
You also need some memory.

#### WBahn

Joined Mar 31, 2012
25,913
So the question still stands -- what IS the hardest question that is reasonable to expect an applicant for an EE position to be able to solve under the conditions of an interview?
How about a problem like this:

Given the following items:
1) A good 1.5 V D-cell
2) A good 1.5 V flashlight bulb
3) A one-foot long insulated wire with the insulation stripped back one inch on each end.

Problem: Using only these components, make the bulb light up.

Should it be reasonable to expect a new grad with an EE degree to be able to accomplish this task under the conditions of a job interview?

#### WBahn

Joined Mar 31, 2012
25,913
A simple XOR should do the trick
Not seeing how that would work. Presumable, in most cases, eventually all three horses will have crossed the finish line in their lane. So given only that information, how can you tell which one crossed first. You are going to need a memory element to capture which horse was first plus a means of ensuring that that information is not corrupted as other horses cross later.

#### crutschow

Joined Mar 14, 2008
25,117
Should it be reasonable to expect a new grad with an EE degree to be able to accomplish this task under the conditions of a job interview?
Probably.
But you are pushing it.

#### WBahn

Joined Mar 31, 2012
25,913
Probably.
But you are pushing it.
Sadly, it was apparently unreasonable to expect MIT EE grads to be able to do it on graduation day.