OTG DIY assembly problem (USB)

This Forum is about the Opentherm gateway (OTGW) from Schelte

Moderator: hvxl

Post Reply
etmrugl
Starting Member
Starting Member
Posts: 2
Joined: Sat Dec 20, 2014 9:42 pm

OTG DIY assembly problem (USB)

Post by etmrugl »

Hi,

I have a OTG DIY USB. I started the measurements according to http://otgw.tclcode.com/debugging.html#power

steps to follow:
The first thing to check is the power supply. Start out with the gateway unplugged from mains power, thermostat and boiler disconnected, and the IC's removed from their sockets. Now measure the resistance between the two terminals of X3. This should be in the order of kΩs.
My value is 3.066 kOhms
Check for 5V +/- 1% between pins 5 and 14 of IC1
I measure 6,67 V. The polarity is correct. The voltage starts at 6,51 and slowly rises

I thought the 7805 was at fault and replaced it. No luck. The input of the 7805 is stable at 24,7 V DC. Next I checked C1 (272 microF) and C2 (276 microF)

My conclusion would be that C2 is damaged, is that correct? (I am completely new to electronics)

Ruud.
hvxl
Senior Member
Senior Member
Posts: 1965
Joined: Sat Jun 05, 2010 11:59 am
Contact:

Re: OTG DIY assembly problem (USB)

Post by hvxl »

I don't assume you need me to tell you that 6.67V is not correct. You'll need to figure out where that comes from. A defective 7805 was a good first suspect. But unless you have multiple bad ones, that doesn't appear to be the culprit. If the first one survived the extraction from the PCB, you could check it outside the circuit. Hook up a 9V power supply or 9V battery to the input and verify that the output is 5V.

If the 7805 is OK, there could be a high frequency signal that manages to get through it, or a short-circuit from the higher voltage part to some component in the lower voltage part. If you ever remove the 7805 again, check the 5V without IC3 in place. If you get a reading above 0 then, there's likely a short circuit somewhere. It sounds like you have a decent multi-meter. It should indicate a very low AC voltage when you measure the input to the 7805.

Just to cover all the bases: Are you sure you're measuring at the correct pins of the correct IC? Alternatively you may check for 5V between pin 15 and 16 of IC2, or pin 1 of the GPIO header and pin 1 of the LEDS header.
Schelte
etmrugl
Starting Member
Starting Member
Posts: 2
Joined: Sat Dec 20, 2014 9:42 pm

Re: OTG DIY assembly problem (USB)

Post by etmrugl »

I tested the old 7805 (the one I thought was faulty and I replaced) Hooked up to a 9 V battery the output is 4.983 V.
I am pretty sure that the voltage is too high, on IC2 pin 15 and pin 16 and pin 1 of the GPIO header and pin 1 of the LEDS header I measure 6.4 V.

So I took out the 7805 again and damaged the board beyond repair. too bad.

Ruud
honeywell
Starting Member
Starting Member
Posts: 2
Joined: Tue Dec 23, 2014 2:29 am

Re: OTG DIY assembly problem (USB)

Post by honeywell »

Hello,

I had the exact same problem. This may be due to the fact that a low power (L = 100mA) version of the 7805 was shipped with the kit that I ordered from http://www.opentherm-gateway.com. After replacing it with a regular 7805 (1A) everything works as expected. That said, it is supposedly bad practice to have a very high difference between input and output voltage of a 7805 voltage regulator - and the design of the gateway is such that it *is* high (24V input, which is close to the maximum: http://www.ti.com/product/ua7805?CMP=AFC-conv_SF_SEP - I read somewhere that it advised is to have about 3V difference, which would be 8V)

Regards,
Tom
hvxl
Senior Member
Senior Member
Posts: 1965
Joined: Sat Jun 05, 2010 11:59 am
Contact:

Re: OTG DIY assembly problem (USB)

Post by hvxl »

That's all very well when designing a power supply for a device that will only have a 5V part, but the otgw needed a 24V part as well. Since not a lot of power was needed on the 5V side, the 7805 wouldn't have to dissipate too much heat. It would have been much more cumbersome and expensive to design an additional 8V power supply just so the 7805 would be most comfortable to deliver those few milliamps at 5V.

But you raise an interesting point regarding the L version. The data sheets I've seen are not very clear on the subject. They only specify that anything above 35V may damage an 78(L)05. Reading between the lines it seems that an 7805 should work OK up to an input voltage of 25V, while the L version may start to misbehave when it gets more than 20V. That would make an 78L05 unfit for this application.
Schelte
Nightlens1
Starting Member
Starting Member
Posts: 13
Joined: Wed Mar 04, 2015 11:07 pm

Re: OTG DIY assembly problem (USB)

Post by Nightlens1 »

I appear to have the same problem: between pin 5 and 14 of IC1I measure 6V instead of 5V.

The in-voltage of the voltage regulator (IC3) is 23.1V and the out-voltage again 5.9V. I tried to remove IC3 but found this to be very difficult. Instead I connected IC3 to my powersource. Between 9-16 in-volt, the out-voltage is 5V. unfortunatly my power source doesnt go higher than 16V, so I cant test 23V.

I just ordered the OTGW kit at KIWI electronics. Several of the components I received were not identical to the ones one their site, including the voltage regulator as I received L7805CV instead of TO220 7805TV.

Any help would be very much appreciated :)
hvxl
Senior Member
Senior Member
Posts: 1965
Joined: Sat Jun 05, 2010 11:59 am
Contact:

Re: OTG DIY assembly problem (USB)

Post by hvxl »

Have you contacted the webshop and told them that they didn't deliver the exact components they advertised and that the L7805CV they sent as a replacement for the 7805TV is not fit for this application? With a bit of luck they will provide the correct part to you.

Alternatively you can get a 7805TV yourself. In either case the easiest way to replace the wrong component is probably to cut the legs of the L7805 and then remove them one by one with a soldering iron and a pair of pliers.

If you don't want to remove the L7805, another possible solution may be to lower the input voltage of the regulator by putting a zener in front of it. Depending on what you connect to the 5V part, the zener should be able to handle around 50mA. Regular zeners are usually rated 500mW, so get one below 10V. Going this route would involve cutting two print traces and soldering the zener to the bottom of the circuit board. I haven't tested if this solution actually works, but if you want to try it and need more details, let me know.
Schelte
Nightlens1
Starting Member
Starting Member
Posts: 13
Joined: Wed Mar 04, 2015 11:07 pm

Re: OTG DIY assembly problem (USB)

Post by Nightlens1 »

Thanks for the advice.
Got in contact with the supplier. Will report how it will end
PeterP
Starting Member
Starting Member
Posts: 5
Joined: Wed May 13, 2015 4:31 pm
Location: Netherlands

Re: OTG DIY assembly problem (USB)

Post by PeterP »

Investigated this problem somewhat more. Can it be that you mixed Q4 and Q5 (BC548 versus BC558)? You easy can make this mistake mainly due to the PCB layout. It will result in a too high +5 voltage behind the 7805 stabilizer.
In case of mix-up: exchange Q4 and Q5 with new transistors of the same type. Measure +5v voltage (between pin 5 and 14) before re-installing IC1.
hvxl
Senior Member
Senior Member
Posts: 1965
Joined: Sat Jun 05, 2010 11:59 am
Contact:

Re: OTG DIY assembly problem (USB)

Post by hvxl »

Can you elaborate on how mixing up Q4 and Q5 could result in a too high +5V? Or did you mean mixing up Q1 and Q4? But then I would expect VCC to end up much higher than the 6 or 7 volts reported by etmrugl and Nightlens1. Although I have to admit I'm not entirely sure what the behavior of the 7805TV would be in that case and how it might influence that value.
Schelte
PeterP
Starting Member
Starting Member
Posts: 5
Joined: Wed May 13, 2015 4:31 pm
Location: Netherlands

Re: OTG DIY assembly problem (USB)

Post by PeterP »

@ Schelte
Sorry for my delay in reaction.
Received one unit which did have the mix-up of Q4 and Q5. In practice it did result in a low 24v (think around 18v) and a too high voltage on pin 14 of IC1. After removing the voltage stab (IC3), I still did have the high voltage on pin 14 (without IC1 installed). As an EE I can't explain why in this position a mix of pnp/npn does result in this situation but after fix it did work again.
PeterP
Starting Member
Starting Member
Posts: 5
Joined: Wed May 13, 2015 4:31 pm
Location: Netherlands

Re: OTG DIY assembly problem (USB)

Post by PeterP »

@ Schelte.
Sorry for the technical explanation:
BC5xx specs do show a VBE0 voltage of 5v (applicable to wrong Q5). For Q4, exchanging an npn transistor with an pnp transistor will give you a max VCB voltage of 0.7v (diode voltage). In case of Q4/Q5 mix-up, together this will result in a kind of zenerdiode of around 5.7v between the 24v and 5v lines. Due to the limited capacity of the powersupply, this will result in a pull-down of the 24v in combination with a rise of the 5v (even without IC3 installed). Hope this explains it well enough.
hvxl
Senior Member
Senior Member
Posts: 1965
Joined: Sat Jun 05, 2010 11:59 am
Contact:

Re: OTG DIY assembly problem (USB)

Post by hvxl »

Interesting. Thanks for taking the time to explain.
Schelte
Post Reply

Return to “Opentherm Gateway Forum”