ULN2803A losing voltage

[Dougal]

I am trying to use the ULN2803A relay driver for the first time. I am following the set-up as here: http://computerchristmas.com/christmas/link-how_to/HowToId-17/Control_using_a_ULN2803A

I am waiting for acceptance to their forum but, hope to get an answer here faster.

I have connected all inputs with the resistors in series and have wired three resistors on three of the outputs with their other side of their coils to the positive supply (five volts in my case as I aim to use 5 volt dc relays.

When I measure the voltages across each coil in turn it is down from five volts to around 3.3 volts.

Is there a good reason for this? Perhaps I can't use 5 volts as a supply for the relays?

I have the 0 volts from the relay supply connected to pin 9 of the ULN2803A as well as the 0 volts from the TTL signals.

Thanks in advance or in fact if someone else has a schematic other than I have described above showing everything needed to drive 5 volt relays and the chip, I would be most grateful.

 

[BobK]

Why are you putting resistors on the outputs? It is probably these resistors that are dropping the voltage.

Bob

 

[shrtrnd]

Don't know your experience level.
ARE your 5V relay coils, 500mA draw or less, as called for in your circuit?
(If the current draw is excessive, it'll screw you up).
I assume you know the ULN2803A is a transistor array, you lose .7v right there.
If you're using the SIP (Single In-line Package) resistor network called for, are you
SURE you wired it in correctly? (Some SIP resistor networks are configured differently
than others). Is resistor network the right value?
If somebody drew the cirucit, it probably works.
Did you try the relays, or just reading the volt meter? Maybe your volt meter doesn't
read low voltages very well (like the Harbor Freight junkers).
Just some things to look at while you're waiting to get into the other forum.
Good luck with it. It's always something simple we just overlook, or assume is right;
only to find out later we assumed something that we shouldn't have).

 

[Dougal]

Thanks for the replies guys.
I replaced the resistor SIP package with individual resistors of the same value in series with each input. I placed resistors equivalent to the relay coil value in the outputs as the relays are not available to test. I worked out the value according to the current drawn in the relay spec from manufacturers data sheet.
I am now going to go for 24 volt relays with 1200 ohm coils instead because I already have a 24 volt 10A supply in my set-up powering other things so, I'll just take power off that. I can adjust the output of the supply slightly to make up for voltage drops if necessary.
One last thing, If I get all this working properly, do I need anything across the 24 volt supply at pins 9 & 10 to counter back EMF?
Thanks again for your input.

 

[Dougal]

24 volt relays work

I have tried a 24 volt relay and it is switching fine.
Referring to the 4.7k resistors I have used at the input, are they simply providing a form of insurance to the input to help prevent overload of the TTL signal supply. Do I need them? I have seen other schematics where there are no resistors on the inputs.

I had been thinking of using a Zener across the supply voltage, pins 9 & 10 but, have opted for diode/suppression packages available from RS Components to fit the relay coils that I am using. Again I am presuming I won't need the Zener if I use these instead?

Well if you guys don't see any obvious problems now, I will proceed as planned.

Thanks again for your inputs.

To answer the question about my experience - I have been involved with scientific equipment for the best part of 30 years but, have never concentrated on pure electronics. My work has always demanded compromize between electrical, electronic and some mechanical knowledge. As such I have to ask questions of others more knowledgeable now and then to keep me right. Please forgive any simple questions that I may ask, I simply want to do things correctly.

 

[Resqueline]

To quote the datasheet: "The ULN2803A has a 2.7-kΩ series base resistor for each Darlington pair for operation directly with TTL or 5-V CMOS devices."
Also: "The device consists of eight npn Darlington pairs that feature high-voltage outputs with common-cathode clamp diodes for switching inductive loads."

So, there's no reason for you to add any resistors or diodes anywhere, as long as you drive them from 5V TTL and connect pin 10 to the relay positive supply..
Decoupling cap's will take care of any spikes, not zeners. The extra 4.7k input resistors were probably the cause of the ULN's failure to drive the 5V relays.

 

[Dougal]

Thanks Resqueline.