10W linear amplifier #problem #linear


Kevin Polston
 


Hello,

I recently built the 10W linear amplifier.   The unit powered up ok drawing 40mW.  I was able to bias it with no problems.  However, when supplied with a sine input of 2Vpp (1MHz) the output was only about 3Vpp also the output “looked" (I’m only a beginner) distorted being elongated and distorted about the mid-point.  The output increased only slightly as I increased the frequency.  Also the unit ran very hot - drawing nearly 3A.  I had check all resistors/caps/transistors and they were in their correct positions.  Then it stopped working - my power supply auto cut-out at over 3A.

I have ordered another and hope to have better luck.   I wondered if anybody could give me a few tips?  I’m guessing that I may have wound a toroid incorrectly? 

The manual is excellent - the only question I have regarding the manual is the wiring of the input/output pads - I guess this is obvious to most but I’m only a beginner - I wondered if anybody could describe/sketch  that wiring?

Many thanks 73

Kevin M0XYM - a rank beginner.
#problem #linear


ajparent1/KB1GMX
 

Obviously made some error that resulted in frying the finals best guess.

As the the question of the IN and OUT pads is there a page reference?

Allison
-------------------------------
Please reply on list so we can share.
No private email, it goes to a bit bucket due address harvesting


ajparent1/KB1GMX
 

When I see somthing like this it begs the queston...

What is the intended use for the amplifier?

Allison
-------------------------------
Please reply on list so we can share.
No private email, it goes to a bit bucket due address harvesting


Kevin Polston
 

Hello Allison,

Thanks for replying to my question. I have no ‘real’ intended use.  I’m just trying to learn RF electronics and the amplifier seemed like a good project - although the design way over my head.  However, I’m also building a small CW transceiver called a Norcal40 and trying educating myself as I build.  When I complete the build of the Norcal40 I might try to combine the two (with a low pass filter of course).  If you have any tips - please don’t say "give-up” - they would be most gratefully received.

Kevin M0XYM

On 29 Sep 2020, at 17:09, ajparent1/KB1GMX <kb1gmx@...> wrote:

When I see somthing like this it begs the queston...

What is the intended use for the amplifier?

Allison
-------------------------------


Kevin Polston
 

Hello Allison,
Yes, - I think you are spot on.  It's the “some error” bit that foxes me.  I perhaps wondered if there is a post-mortem procedure I could attempt that would lead to the cause?  If this is a naive question then please just ignore me - I will get the message.

Regarding the in/out pads.  They are referenced in the (superb) manual but I think the designers didn’t expect a customer with my lack of expertise and confidence so are not “spelt-out".  I had a go, but an interface sketch would have calmed my nerves!
Thanks
Kevin

On 29 Sep 2020, at 17:07, ajparent1/KB1GMX <kb1gmx@...> wrote:

Obviously made some error that resulted in frying the finals best guess.

As the the question of the IN and OUT pads is there a page reference?

Allison


Michael N6MST
 

On Tue, Sep 29, 2020 at 10:02 AM, Kevin Polston wrote:
I’m also building a small CW transceiver called a Norcal40 and trying educating myself as I build.
If you haven't already, make sure to go through the Norcal 40A video series Chuck Adams K7QO put on YouTube.


Kevin Polston
 

Hello Micheal,

Yes, I have followed Chuck Adams Norcal40a series.  The series really helped me - especially his incremental build/test approach.  Unfortunately the final parts of the series are missing (I think his hard drive crashed and he lost them).  I also got a copy of a book called the “Electronics of Radio” by David B. Rutledge which is currently my text book and bible for learning RF electronics - it uses the Norcal40a.

All the best
Kevin M0XYM
PS I haven’t heard much activity from Chuck in the past few months - I hope he’s ok...

On 29 Sep 2020, at 18:05, Michael N6MST <mthompson.cob@...> wrote:

On Tue, Sep 29, 2020 at 10:02 AM, Kevin Polston wrote:
I’m also building a small CW transceiver called a Norcal40 and trying educating myself as I build.
If you haven't already, make sure to go through the Norcal 40A video series Chuck Adams K7QO put on YouTube.


Michael N6MST
 

On Tue, Sep 29, 2020 at 10:21 AM, Kevin Polston wrote:
I haven’t heard much activity from Chuck in the past few months - I hope he’s ok...
Last I heard he was stepping away from radio projects and he shut down his qrp-tech group to focus on other endeavors. He seems to have kept at radio at least a little bit since then and he is still updating his k7qo.com website. My guess is he's doing fine :)


ajparent1/KB1GMX
 

Well the some error is based on scanty info so many things are possible
and likely.

ITs a linear amp but not the traditional style of one power amp stage 
so it has mroe gain and lower powerinput requirements/tolerance.

As to using it for an amp for the norcal...  First you have to drop the N40
to about 50mW output or you fry the amps input.  Then a LPF for 40M.
Then T/R switching so the amp is not in the path during receive.

The N40 is 3-5W and to make sense at all you need to double your
power out to make it worth while and so the other guy can see it.
Generally from QRP the next best power is 20W.

Allison
-------------------------------
Please reply on list so we can share.
No private email, it goes to a bit bucket due address harvesting


Kevin Polston
 

Allison,

All noted - especially your caution that that the amp mustn’t be in the receive path.  A sincere thanks for taking the trouble to reply.  I now realise that I have chosen a project (the amp) that is considerably above my level of competence and understanding.  However, I am determined to sucessfully build it even if it’s mode of operation eludes me for some time.

Thanks again
Kevin

On 29 Sep 2020, at 18:25, ajparent1/KB1GMX <kb1gmx@...> wrote:

Well the some error is based on scanty info so many things are possible
and likely.

ITs a linear amp but not the traditional style of one power amp stage 
so it has mroe gain and lower powerinput requirements/tolerance.

As to using it for an amp for the norcal...  First you have to drop the N40
to about 50mW output or you fry the amps input.  Then a LPF for 40M.
Then T/R switching so the amp is not in the path during receive.

The N40 is 3-5W and to make sense at all you need to double your
power out to make it worth while and so the other guy can see it.
Generally from QRP the next best power is 20W.

Allison
-------------------------------
Please reply on list so we can share.
No private email, it goes to a bit bucket due address harvesting


Curt wb8yyy
 

Kevin

you have excellent advisors here already - and glad you are joining us in building. 

you should spend a little time seeing what if anything went wrong with the linear PA.  note being linear it draws DC current without RF being fed into it (unlike some other examples like the QCX PA that require input RF to draw much current). 

any visual damage?  if you want to check the mosfet final stages, you may need to disconnect a few parts - but if have had a similar device (okay I doubt you may have one) it could be compared to a fresh device.  often a serious issue involves too much current flow, then some parts may be 'smoked' so look carefully. 

enjoy a patient build of that norcal40 - and do seek Allison's advice later in selecting a PA for it (like maybe the QCX PA with a mod to provide the TR control).  possibly the 10w linear amplifier could be repaired from parts from another amateur's 'junk box.'  this amplifier is intended for a future rig called the QSX.  if you get interested in modes like WSPR - the U3S is a fun device and its doesn't need an amplifier to cross oceans.  do exercise the community here - glad to have you with us. 

Curt wb8yyy


geoff M0ORE
 

Hi Kevin.

The 10 watt PA module is intended as part of the QSX SSB transceiver when it is released. As it is a linear  amplifier, it is not really suitable for a CW only rig. It will work but not very efficiently.

I see that you are a member of the Chelmsford RS, a club with a long history serving the amateur fraternity.  I suggest that you ask some of the members of the excellent club that you are a member of for assistance.



On 29/09/2020 19:11, wb8yyy via groups.io wrote:
Kevin

you have excellent advisors here already - and glad you are joining us in building. 

you should spend a little time seeing what if anything went wrong with the linear PA.  note being linear it draws DC current without RF being fed into it (unlike some other examples like the QCX PA that require input RF to draw much current). 

any visual damage?  if you want to check the mosfet final stages, you may need to disconnect a few parts - but if have had a similar device (okay I doubt you may have one) it could be compared to a fresh device.  often a serious issue involves too much current flow, then some parts may be 'smoked' so look carefully. 

enjoy a patient build of that norcal40 - and do seek Allison's advice later in selecting a PA for it (like maybe the QCX PA with a mod to provide the TR control).  possibly the 10w linear amplifier could be repaired from parts from another amateur's 'junk box.'  this amplifier is intended for a future rig called the QSX.  if you get interested in modes like WSPR - the U3S is a fun device and its doesn't need an amplifier to cross oceans.  do exercise the community here - glad to have you with us. 

Curt wb8yyy


Bill Cromwell
 

Hi,

I have and use radios built for CW and SSB and they have linear output stages that serve without complaint in the CW mode. If it is what you have and it works just use it.

73,

Bill KU8H

bark less - wag more

On 9/29/20 7:06 PM, geoff M0ORE via groups.io wrote:
Hi Kevin.
The 10 watt PA module is intended as part of the QSX SSB transceiver when it is released. As it is a linear  amplifier, it is not really suitable for a CW only rig. It will work but not very efficiently.
I see that you are a member of the Chelmsford RS, a club with a long history serving the amateur fraternity.  I suggest that you ask some of the members of the excellent club that you are a member of for assistance.
On 29/09/2020 19:11, wb8yyy via groups.io wrote:
Kevin

you have excellent advisors here already - and glad you are joining us in building.

you should spend a little time seeing what if anything went wrong with the linear PA.  note being linear it draws DC current without RF being fed into it (unlike some other examples like the QCX PA that require input RF to draw much current).

any visual damage?  if you want to check the mosfet final stages, you may need to disconnect a few parts - but if have had a similar device (okay I doubt you may have one) it could be compared to a fresh device.  often a serious issue involves too much current flow, then some parts may be 'smoked' so look carefully.

enjoy a patient build of that norcal40 - and do seek Allison's advice later in selecting a PA for it (like maybe the QCX PA with a mod to provide the TR control).  possibly the 10w linear amplifier could be repaired from parts from another amateur's 'junk box.'  this amplifier is intended for a future rig called the QSX.  if you get interested in modes like WSPR - the U3S is a fun device and its doesn't need an amplifier to cross oceans.  do exercise the community here - glad to have you with us.

Curt wb8yyy


Kevin Polston
 

Bill, Curt, Geoff

All - Thanks for the advice.

@Geoff - Yes, Chelmsford ARS is indeed a fine club - but the trouble is with this pesky virus that meetings are suspended for the foreseeable future.

@Curt - No, visible damage.  I’m ordering a second amp - hopefully I’ll get that working.  Also, I’ve ordered some replacement IRF510’s for my original unit.  As my experience increases maybe I’ll eventually work-out how I killed-it and having a working bit (hopefully) will assist in the post-mortem.

Kevin


On 30 Sep 2020, at 14:48, Bill Cromwell <wrcromwell@...> wrote:

Hi,

I have and use radios built for CW and SSB and they have linear output stages that serve without complaint in the CW mode. If it is what you have and it works just use it.

73,

Bill  KU8H

bark less - wag more

On 9/29/20 7:06 PM, geoff M0ORE via groups.io wrote:
Hi Kevin.
The 10 watt PA module is intended as part of the QSX SSB transceiver when it is released. As it is a linear  amplifier, it is not really suitable for a CW only rig. It will work but not very efficiently.
I see that you are a member of the Chelmsford RS, a club with a long history serving the amateur fraternity.  I suggest that you ask some of the members of the excellent club that you are a member of for assistance.
On 29/09/2020 19:11, wb8yyy via groups.io wrote:
Kevin

you have excellent advisors here already - and glad you are joining us in building.

you should spend a little time seeing what if anything went wrong with the linear PA.  note being linear it draws DC current without RF being fed into it (unlike some other examples like the QCX PA that require input RF to draw much current).

any visual damage?  if you want to check the mosfet final stages, you may need to disconnect a few parts - but if have had a similar device (okay I doubt you may have one) it could be compared to a fresh device.  often a serious issue involves too much current flow, then some parts may be 'smoked' so look carefully.

enjoy a patient build of that norcal40 - and do seek Allison's advice later in selecting a PA for it (like maybe the QCX PA with a mod to provide the TR control).  possibly the 10w linear amplifier could be repaired from parts from another amateur's 'junk box.'  this amplifier is intended for a future rig called the QSX.  if you get interested in modes like WSPR - the U3S is a fun device and its doesn't need an amplifier to cross oceans.  do exercise the community here - glad to have you with us.

Curt wb8yyy 




Timothy Fidler
 

TEF has been in contact with  Kevin with about 5 suggestions as to what he may have done.  I went direct because the moderation bizzo that has been visited upon me like the plage of LICE,  is a Pita.  

The most obvious Fox's pas he might have made is to   make up the  T 203 , ie the phasing trafo  with    a  - a' and b - b' bifilar wire runs  and mistakenly feed VCC down a and b starts (as a node )  rather than to the essential symmetry cross over.  The build book goes into this extensively - precisely what to do.  With no drive you would see not much ill effect but with RF drive you'd short the MOSFETs (IRFs) into the phasing transformer.   I suggested on go around mark two he use an indelible black texter to mark wire  b-b' of the bifilar pair so it is impossible to mix up what lead goes where.. Clearly it is an arbitrary decision as to which physical bit of copper you make black.

That of course might not be the fellow in the woodpile in this occasion.   But if it is then the black wire trick might stop repeats - I 'll get K to note this to Hans if indeed it is the fix. 

I asked him to check on how this transformer was indeed soldered in but have not heard back.

As for the Rx Tx relay and change over time .. I copied Kevin in on the build book for the TexasTopper  cheap and dirty  Class  C RF amp out of the USA . I told him he can do a heist  on the  Carrier operated relay from there for his use.  sometimes it is better not to re invent the wheel.  The TT also shows a LPF implementation though you can't  do a cooky cutter job on that because of the different impedances  between the two amps. 

I think the guy who did the bias circuit on the TexasTopper most likely also worked on the famous cryo stir bottle on Apollo 13. Looks that way.  The bias LED fails and the TT is under RF drive - bang, the Mosfet blows. Now that's class.




Kevin Polston
 

Hello Timothy,
I have made a royal mess of trying to desolder the transformers.  And as a consequence have ordered a solder sucker - not the cheapest but not the dearest either (a Japanese model with silicon tips for about GBP15 from Amazon).  I have a sneaking suspicion that I may have mis-wired the T203 as you say - although I *thought* I was following the instructions - but maybe not. My new amp should be here soon and I am going to be more careful and deliberate.  It’s a shame I won’t be able to diagnose what really went wrong but I have enough problems understanding things when they work properly!
Looking forward to building my new unit but I am going to take time.

Thanks

On 1 Oct 2020, at 23:37, Timothy Fidler <engstr@...> wrote:

TEF has been in contact with  Kevin with about 5 suggestions as to what he may have done.  I went direct because the moderation bizzo that has been visited upon me like the plage of LICE,  is a Pita.  

The most obvious Fox's pas he might have made is to   make up the  T 203 , ie the phasing trafo  with    a  - a' and b - b' bifilar wire runs  and mistakenly feed VCC down a and b starts (as a node )  rather than to the essential symmetry cross over.  The build book goes into this extensively - precisely what to do.  With no drive you would see not much ill effect but with RF drive you'd short the MOSFETs (IRFs) into the phasing transformer.   I suggested on go around mark two he use an indelible black texter to mark wire  b-b' of the bifilar pair so it is impossible to mix up what lead goes where.. Clearly it is an arbitrary decision as to which physical bit of copper you make black.

That of course might not be the fellow in the woodpile in this occasion.   But if it is then the black wire trick might stop repeats - I 'll get K to note this to Hans if indeed it is the fix. 

I asked him to check on how this transformer was indeed soldered in but have not heard back.

As for the Rx Tx relay and change over time .. I copied Kevin in on the build book for the TexasTopper  cheap and dirty  Class  C RF amp out of the USA . I told him he can do a heist  on the  Carrier operated relay from there for his use.  sometimes it is better not to re invent the wheel.  The TT also shows a LPF implementation though you can't  do a cooky cutter job on that because of the different impedances  between the two amps. 

I think the guy who did the bias circuit on the TexasTopper most likely also worked on the famous cryo stir bottle on Apollo 13. Looks that way.  The bias LED fails and the TT is under RF drive - bang, the Mosfet blows. Now that's class.





Henning Weddig
 

Timothy,

I also had trouble to get the amp into operatation: the cause was the wrong winding of T203. As I remember the gain at low frequency was bad.  After correcting the windings the amp works.

Hans: for the next revision of the amp please enlarge the holes and pads for the "thick wires" of the transformers, in order to ease possible errors and folllowing desoldering actions. Also the manual should be made clearer how the windings have to be inserted.

73

Henning Weddig

DK5LV

Am 02.10.2020 um 10:29 schrieb Kevin Polston via groups.io:

Hello Timothy,
I have made a royal mess of trying to desolder the transformers.  And as a consequence have ordered a solder sucker - not the cheapest but not the dearest either (a Japanese model with silicon tips for about GBP15 from Amazon).  I have a sneaking suspicion that I may have mis-wired the T203 as you say - although I *thought* I was following the instructions - but maybe not. My new amp should be here soon and I am going to be more careful and deliberate.  It’s a shame I won’t be able to diagnose what really went wrong but I have enough problems understanding things when they work properly!
Looking forward to building my new unit but I am going to take time.

Thanks

On 1 Oct 2020, at 23:37, Timothy Fidler <engstr@netspace.net.au <mailto:engstr@netspace.net.au>> wrote:

TEF has been in contact with  Kevin with about 5 suggestions as to what he may have done.  I went direct because the moderation bizzo that has been visited upon me like the plage of LICE,  is a Pita.

The most obvious Fox's pas he might have made is to   make up the  T 203 , ie the phasing trafo  with    a  - a' and b - b' bifilar wire runs  and mistakenly feed VCC down a and b starts (as a node )  rather than to the essential symmetry cross over.  The build book goes into this extensively - precisely what to do.  With no drive you would see not much ill effect but with RF drive you'd short the MOSFETs (IRFs) into the phasing transformer.   I suggested on go around mark two he use an indelible black texter to mark wire  b-b' of the bifilar pair so it is impossible to mix up what lead goes where.. Clearly it is an arbitrary decision as to which physical bit of copper you make black.

That of course might not be the fellow in the woodpile in this occasion.   But if it is then the black wire trick might stop repeats - I 'll get K to note this to Hans if indeed it is the fix.

I asked him to check on how this transformer was indeed soldered in but have not heard back.

As for the Rx Tx relay and change over time .. I copied Kevin in on the build book for the TexasTopper  cheap and dirty  Class  C RF amp out of the USA . I told him he can do a heist  on the  Carrier operated relay from there for his use.  sometimes it is better not to re invent the wheel. The TT also shows a LPF implementation though you can't  do a cooky cutter job on that because of the different impedances  between the two amps.

I think the guy who did the bias circuit on the TexasTopper most likely also worked on the famous cryo stir bottle on Apollo 13. Looks that way.  The bias LED fails and the TT is under RF drive - bang, the Mosfet blows. Now that's class.



Timothy Fidler
 

for plage read Plague (of Lice)  TEF


Rod Smith
 

Life, TEF, is indeed a beach ;0)