开云体育

On Fri, Oct 26, 2018 at 11:22 AM, Jerry Gaffke wrote:
With the PTT switch pressed, measure the following voltages:
? ? TX? ? (should be +12v)
? ? U3 pin 1? ?(should be +5v)? ?Could alternately be measured at RV1 or C150
? ? R150? ? (both sides the same and same as the IRF510 gate, should vary between 0v and 5v as you turn RV1).

All of the voltages above are DC, measured with the black probe on ground a

Is this a bitx40 or a uBitx???? They are very different and I keep seeing references to both!

++++++++++++++++++++++++++++++++++++? uBitx only!++++++++++++++++
for ubitx U2 is the 5V source for get bias.? RV2 and RV3 are the bias adjust for each IRF510.

Note no audio into the mic or use a jumper to connect PTT to ground for SSB TX.

To dial in the bias urn the pots RV2 and 3 down to 0V on the gate (oddly its clockwise!).
Measure the radio current, whatever that is adjust RV2 for 100mA more.? The resulting
current is noted and then adjust RV3 for an additional 100ma.? Total current will be about
350mA give or take depending on the starting number.

With 12V on the drains the gates should be at 3-4V and about the same voltage for each
IRF510.

IF the gates are still at zero volts then you have other issues with burned traces.

That being done RV1 set to about 10 o-clock position (2/3 power) (its also backward)
as max power out is anti-clockwise.? Now you should see some power out.

Allison

I suggest you pull the two pin PA-PWR1? connector from your Bitx40
while debugging this gate voltage issue.
With no 12v power to the IRF510 drains, minor explosions are much less likely.

You will only see a DC voltage on the IRF510 gate when transmitting, with the PTT switch closed.
The PTT switch grounds the bottom of the coil at relay K1, turning K1 on.
With K1 on, we have the +12v on K1 pin 5 connected to the TX node on K1 pin 1.
The +12v on TX goes into pin 3 of the LM78L05 regulator at U3, which should now provide +5v on pin 1.
The 10k pot at RV1 can be adjusted to provide any voltage between 0v and 5v at RV1 pin 2, we will adjust
it eventually to be roughly 4v.
Note that RV1 pin 2 is not connected to the LM78L05 at U3, it is just that the schematic is poorly drawn,
instead that RV1 pin 2 node goes down the schematic to the 47 ohm resistor at R150.
Since there is no DC load of any sort on the bottom of R150, the bottom (which is connected to the IRF510 gate)
should be at the same voltage as the top of R150.? So varies between 0 and 5v as you turn RV1.

With the PTT switch pressed, measure the following voltages:
? ? TX? ? (should be +12v)
? ? U3 pin 1? ?(should be +5v)? ?Could alternately be measured at RV1 or C150
? ? R150? ? (both sides the same and same as the IRF510 gate, should vary between 0v and 5v as you turn RV1).

All of the voltages above are DC, measured with the black probe on ground and the red probe on the signal of interest.
Your voltmeter should be set to perhaps 20 volts max, higher than your power supply voltage.
Make sure the meter is configured to measure voltage and not Amps.
If you get the IRF510 gate voltage to vary with RV1, then turn RV1 fully clockwise for minimum gate voltage.
This the opposite direction for minimum of most pots.

Now change your meter from volts to measure amps instead (at least 1 Amp max), and connect it in series with your power supply +12v wire
that goes into the Bitx40.? With the power supply off, reconnect the PA-PWR1? two pin connector to the Bitx40 board.
Put a 10W (or more) 50 ohm dummy load on the antenna port of the Bitx40.?
(You do have a dummy load, don't you?? If not, I'd recommend four Xicon 200 ohm 3W in parallel, Mouser?283-200-RC.)

Get some safety glasses on, as those IRF510's can explode, blowing out molten metal.
Cross your fingers and power up the supply.
I don't remember exactly what the current draw from your supply should be for a Bitx40, but believe it is around 150 ma.
Now ***very*carefully*** turn RV1 counterclockwise until that that initial power supply current increases by 100 ma (to about 250ma total).
I assume you are in SSB mode (not CW), and? do not have audio into the mike.
At first, turning RV1 won't have any effect, but a bit beyond halfway the current draw will suddenly increase rapidy as you turn.
Turn RV1 too far and you may find out why I recommend safety glasses.

When operating with audio into the mike, the Bitx40 might draw a couple amps from the power supply.

Someone attempting to fix a Bitx40 power amp? that is unsure how to use a voltmeter
is definitely jumping into the deep end of the pool.
If any of the above is confusing, I suggest you check in with a local ham club,?
see if anybody there is willing to help you sort this out.
Bring your Bitx40, schematic, PTT switch, power supply, dummy load, meter, and this post.
They may be able to get you going in the right direction in a few minutes.
?
Jerry

Is there a transmit calibration to do once I add this mod?

No difference with LSB or USB while in transmit ...still the same DDS vfo output signal....terrible distorted audio!

What about changing the Raduino code to reflect the shift in TX?? I am using v1.28.1?

LSB?? 4.81329 MHz DDS output looks like a square wave!

USB?? 19.1917 MHz DDS output

Just detected something else:? When I let go of the PTT the frequency drops100Hz each time!

Anyone know what the purpose of the blue pot on the Raduino...contrast ?

Help...please!

Michael

On Thu, Oct 25, 2018 at 11:21 PM, Glenn wrote:
another small issue..... all holes in the heat-sink have burs on them.?

You want to keep it FLAT as the devices have to be flat to it for best heat transfer.
Burs are easily removed with a precision knife (Exacto or similar).? A?good builder
checks for accidental shorts prior to power up.

FYI the bigger thing is that the FBxx-2402 ferrite have sharp edges on the holes
and can strip enamel from wires while winding.? I use a small cone shaped
Carborundum stone to take the edges of the holes down.

Allison


?

On Thu, Oct 25, 2018 at 10:41 PM, Glenn wrote:
The drilled tapped holes in the heat-sink, seem to be 'inaccurate '.? When mine is assembled, the IRF510's have to be skewed a little to fit. ie not fitting parallel to the edges of the PCB and heat-sink.? It's most easily noticable by looking at the IRF510 screws are not in the centre of the PCB holes. It would appear the hole holding down the BS170's is not centred in the heat-sink. I'm guessing it varies from heat-sink to heat-sink. Not a deal breaker.

THe IRF510 holea re centered on the device solder pads and not the edges of anything.? Its undesirable to put the
power device at the edge of the heatsink as you loose transfer area.

The other I do not see on mine but in any case its not a functional factor and far from a deal breaker.
I see no issue and for the price its a a good value.

If you look on QRPlabs page rather than bitx at the pictures of the QSX strict alignment is not required.

Allison

Are you sure your Bitx40 is working well in receive?
Do you have single signal reception?
So when you tune through a short wave broadcast station,?
you should hear the carrier coming through loud and clear on one side of zero-beat
and hardly at all on the other side.

In answer to your question, the VFO stays put when switching between transmit and receive.

Assume the BFO is at 11998000 hz, and the 2000 hz wide?12mhz crystal passband
is between 11995500 and 11997500 hz.
The upper edge of that passband is 500 hz below the BFO, and the center of that passband is 1500 hz below the BFO.
But the math for calculating the required VFO frequency has only to do with the?
the BFO frequency and the suppressed carrier of the signal, using same math for both transmit and receive.?

If we are receiving an LSB signal at 7200000 hz, then the VFO will be at 11998000 - 72000000 = 4798000 hz.
The 7200000 hz figure is that of the suppressed carrier, not of the 2000 hz wide translated audio that the far transmitter is actually sending out.
The center of that translated audio we have coming in from the ether is at 7200000 - 1500 = 7198500 hz.?
The VFO mixes with this to produce 7198500 + 4798000 =?11996500?hz, which is at the center of our 12mhz filter passband.

The same math applies when we are transmitting with a (suppressed carrier) frequency of 7200000 hz..
The VFO remains at 11998000 - 72000000 = 4798000 hz, and the center of stuff coming out of our 12mhz filter
at 11996500 mixes with the VFO to produce translated audio centered on 7198500 hz, which goes out over the air.
.
When using digital modes, we want the Bitx40 to use USB mode.
Allard's code does this by going to a high side VFO, in this case at 7200000 + 11998000 = 19198000 hz.
When transmitting, the the center of the outgoing translated audio is 19198000 - 11996500 = 7201500 hz.
Note that this is now above the suppressed carrier, not below.

The uBitx is double conversion, making it a slightly more?challenging?puzzle.
See post? ?/g/BITX20/message/44515

When in CW mode, there is a VFO shift between transmit and receive.
This is done so the received carrier gives a roughly 600 hz tone instead?of being at zero-beat.

Jerry, KE7ER


??

Hi Michael,

With the stock raduino firmware, the Raduino doesn't "know" whether the radio is in receive or in transmit (hence it cannot do RIT,SPLIT, etc.)
The new software at ?supports these functions provided that the "PTT-sense mod" is installed.

For PTT-sense you only need to add a 10K resistor as shown here:


73 Allard PE1NWL

My BITX40 sounds good in receive - especially after I installed all new crystals from Digi-Key (tall version).

Anyway transmit is still distorted and hard to make out. On my K3's Panadapter it looks like double sideband.

This got me to think how does the Raduino know it's in transmit?? I think I am missing a wire to two to it? The online
wiring diagram just shows the tuning pot, VFO out, and power.? What about the transmit - shouldn't the VFO or DDS
signal be shifted when in transmit???

What about the PTT sense to A0 shown in the attached diagram?

Thank you,

Michael
n2zdb

Hi folks,

I've messed up this ubitx a lot. Learning as I go... I guess. Today I pulled it out again to give it another shot, and found the burnt trace that caused me to put it away 2 months ago. That wasn't even the original issue with it, so I knew I had to find the other one. I have 12 volts on the tabs of the irf510s, but none on the gates while TXing. IIRC there should be around 4 volts there. Somehow while testing voltages between ground and the IRFs I burned another trace (between PA PWR and L8). Luckily this one was far easier to fix.?

Prior to burning the traces, I came here trying to solve the issue. Here's the link to that thread if it helps at all:?/g/BITX20/topic/24923512?p=,,,20,0,0,0::relevance,,No+RF+out+SSB,20,2,0,24923512,ct=1&ct=1

No idea what the problem is, I've been at this for over 6 months now, on and off of course. Any help figuring this out is much appreciated. Thanks!
-Evan

another small issue..... all holes in the heat-sink have burs on them.?? Depending on how bad they are, they may cut through the insulator and ground the IRF510 tabs. Maybe not now, maybe later.
I sanded my heat-sink smooth before fitting the PCB etc.

vk3pe

Observations on this Kit, mechanical.

I have been assembling this kit today and noticed a couple of things, mechanically, when building. (yet to power up)

The drilled tapped holes in the heat-sink, seem to be 'inaccurate '.? When mine is assembled, the IRF510's have to be skewed a little to fit. ie not fitting parallel to the edges of the PCB and heat-sink.? It's most easily noticable by looking at the IRF510 screws are not in the centre of the PCB holes. It would appear the hole holding down the BS170's is not centred in the heat-sink. I'm guessing it varies from heat-sink to heat-sink. Not a deal breaker.

The use of the large nut under the BS170's means the PCB does not fit parallel to the heat-sink. ie the large nut end is much lower than the IRF510 height. I 'fixed' this by turning up a new spacer from aluminium rod. I drilled the centre hole to 3mm so that the rubber tube spacer is not required.? Height of the spacer is approx 5.7mm and the diameter is 9.5mm. (The supplied nut is ~4.4mm high.) Another 'fix' would be? a suitable flat washer, on the heat-sink side,.

Glenn? vk3pe

Pic shows supplied spacer/nut height is lower than the IRF510's height.


Pic below, new spacer:

Pic below, even height now of the PCB to heat-sink.

Ant is OK and already have an off board audio amp, but thanks.? Plan to add AGC after getting hte stock board working...
Fixing my freq counter moved ahead of mor ubitx work...

Woody -? KZ4AK


On 10/26/2018 1:57, Curt via Groups.Io wrote:

Woody

check to see if you did not mix up the ground and signal connections at the antenna, as I did at first.

also, if you have an outboard device with an audio amplifier, like the NESCAF here or an amplified speaker - see if that can bring up your audio.

I will share that since the rig does not have AGC -- signals can go from being very loud requiring minimum audio gain to very low requiring all the audio gain.

it does take some patience to get this thing wired correctly - don't ask how many issues I had in integrating the uBITX board into the case from . I am hopeful you will find it is something simple to correct (but not simple to find).

73 Curt
_._,_._,_
------------------------------------------------------------------------

--

Woody

check to see if you did not mix up the ground and signal connections at the antenna, as I did at first.?

also, if you have an outboard device with an audio amplifier, like the NESCAF here or an amplified speaker - see if that can bring up your audio.?

I will share that since the rig does not have AGC -- signals can go from being very loud requiring minimum audio gain to very low requiring all the audio gain.?

it does take some patience to get this thing wired correctly - don't ask how many issues I had in integrating the uBITX board into the case from .? I am hopeful you will find it is something simple to correct (but not simple to find).?

73 Curt

A - 24 db electret will likely be enough to have plenty of average power without mods. I use the stock element but have made the one resistor mod at the mic input and for me it works well. The stock element might be cheap but when I tried a 4USD and - 44db I have had less power output. I certainly have had a much worse experience with the IC-7000 stock microphone rather than the allegedly cheap stock ubtix element.

Jim, Thanks for checking your levels. Pretty close to what I have when re-measured (below).

Looks like I will need to do more detective work.? I was misled by my not so great teat equipment.? Visited a RF engineer friend (N4UA) and we checked with his 100 MHz scope. TP 12, 15, & 18 all measured 700 mV P-P. ? Don't know what is normal -? Perhaps lower than I would have guessed, but present and agrees with Jim.

He wondered if the problem may be in the 12 MHz crystal filter. We agreed that a problem with it could cause the symptoms.? I can say that the filter is rather microphonic.?? Lightly tapping the crystals produced sound in the audio output.? Hard to be certain, but 2 of them seemed more sensitive than the others.
Also...
A signal could be heard on 40M with a 1 uV signal to the antenna port.? Not too bad, but the low audio, tuning and narrow bandpass issues remain.

Need to do more signal tracing and check the center, loss, and bandwith of the filter.? Unfortunately, my frequency counter died a few days ago, so will have to? fix it before I can sweep the filter.

So, the uBITX trouble shooting is going to be on hold until I repair my frequency counter.?? I will be back with a solution (or more questions) when the counter is working and I have more measurements.
Thanks again for all the suggestions - Good to have the support!

For now, this tread is on hold....
Woody,? KZ4AK
--

Also, its a large db span, the one I think I've received. And, much smaller. There are huge differences in microphones and Baofeng and perhaps Ashhar are trying to be extremely economical with their choices.? It would make sense for Ashhar to design the microphone amp "around" the microphones he had samples of and when satisfied order a ton of them. I've got a noise generator and scope and thought of designing a measurement jig for some time. And, remembering that the 70 cent one is most likely about 5 cents in quantity.?