US amplifiers.

Do you enjoy making your own equipment or antennas? Discuss construction and design in here.
User avatar
theEarwigger
Super Member
Super Member
Posts: 403
Joined: 24 Jul 2007, 20:59
Call Sign: M0RZF
Location: Portsmouth
Contact:

Re: US amplifiers.

Post by theEarwigger » 02 Jun 2018, 11:49

Yes,

Making high power amps is not sticking a few diodes and "pills" in a 'backy tin and it works. If you want a cheap transistor that can deliver 40W a pair, the OnSemi FQP13N10 is used in many CBs. The best DMOS type I found is the IXTP2R4N50P from IXYS. Ironically the IXYS part is going obsolete but still available from Mouser UK.

LDMOS RF parts cost more but there are some in the £20 each price range if you're going for 100W. I wouldn't waste time struggling with obsolete parts from a dodgy supplier. £40 extra on FETs is nothing compared to the time it takes to do reliable amp design.

Interesting how even 30 years after it's heyday, CB manages to come up with terms like "pill" when they mean transistors :D The "pills" are bipolar transistors when everybody (even at lower frequencies) moved to FETs.

-Rob

User avatar
Unit 148 Mobile
Super Member
Super Member
Posts: 110
Joined: 24 Sep 2010, 20:50
Location: London
Contact:

Re: US amplifiers.

Post by Unit 148 Mobile » 02 Jun 2018, 17:52

Sureshot,

May I add my 2p worth.

Spend a few quid on a couple of books first. Learn about RF semiconductors (there's no such thing as a pill transistors, your doctor prescribes pills) in this classic: https://www.ebay.co.uk/itm/Motorola-RF- ... SwT6pVyLgU
It's old but all the info still stands and it will describe some very simple test circuits, correct board layout and the importance of impedance matching.

Second book is: https://www.ebay.co.uk/itm/RF-Circuit-D ... SwOhda2Nss

If you wish to try your first build for 27Mhz then you could build a class-C amp but use it strictly with your FM transceiver. #26 is correct in what he says and I'll add that you'd receive the "Worked all Neighbors" award if you use it on AM or side-band.

Bias circuits and correct layout of them in relation to the final transistor/s are a subject all unto themselves and are covered in the books above.

The US poster claiming the high drive amplifier can claim high output of 400W with 100W drive is I'm afraid typical of poorly designed amplifiers. Even the old, bipolar designs (2SC2879 Toshiba's included) never required that level of drive. I believe they had a typical gain of around 10db. My Metron amplifier under my post has as you can see 8 x 100W devices in the output for between 500-600W output depending on the band you're on.

A single device power amplifier is relatively easy to build but designing all the control circuitry around it (over-temp, VSWR foldback etc) is what makes a good commercial amplifier and which will take you far more time to get working.

Few of the American amplifier builder know anything about RF design. Most all just copy one another. Have you seen the mess of some of those units. I have a heavily modified X-Force 8 x 2SC2879 Toshiba amplifier I brought back from a shootout around 1987 in Chicago. It had to be completely rebuilt and a proper bias circuit installed before I was let loose with it.

A couple of my mobiles: https://ibb.co/hCiMvy

https://ibb.co/b4dkoJ

Unit 148

wa10
Top Poster
Top Poster
Posts: 1169
Joined: 19 Feb 2008, 00:37

Re: US amplifiers.

Post by wa10 » 02 Jun 2018, 21:29

"Few of the American amplifier builder know anything about RF design"

absolutely true 148,

many just copy davemade amps ( swing boxes ) that never were designed for cleanliness or running on stock voltage, the design is nothing like any conventionally designed amp from helge grandbergs motorola application notes or the toshiba datasheet,

the amps were designed to win keydowns against conventional amps which they will do, a 2pill dave style amp WILL whoop 4pill in a keydown,

datasheet nuthuggers are armed with a little info but not enough, toshiba datasheets only tell half the story, look at other transistors that show output & imd at different supply voltages,

Dave is smart, he understood vcc vs load impedance years ago as anybody that builds tube amps should understand, he knew doubling your supply voltage quadruples your output power when optimally load matched so a moderate increase in vcc can yield much more power,

as you raise vcc optimum load impedance increases, change class & optimal load impedance changes,

dave knew that you can run transistors at higher vcc and load impedance at a reduced duty cycle to get more power without exceeding maximum device current or maximum device dissipation, how much more power ? a hell of a lot more than the datasheet tells us,

that its not just current but current x time that smokes the device,
a fuse will handle many times its amp rating if you pulse current through it so reducing the time the device is turned by going class C allows the gold wires that connect the tabs to the die under the beryllium oxide cap that coined the term pills to handle more current than the datasheet tells us,

you can push a device into deep class C by using negative voltage biasing to reduce on time even further than a grounded base type class C,
a few 55 built amps used a pp3 battery hidden inside the amp to get the negative biasing,

Dave understood what the datasheet did and more importantly what it did not tell you, and that absolute ratings are higher than what the datasheet tells us if reliability and imd are of no concern,

you don't build the same engine for a drag race if you want to race round a track for many laps and a different engine again if you want to go shopping or are the slightest bit concerned about emissions, there's no over fuelling to cool cylinder temps in your shopping trolly,

he built class C drag racing amps with the required for keydowns 2sc2879's that ran on 21v or a tad higher ( 18v max on datasheet ) with transformer turns ratios and tuning ( oh the top secret tuning on 6pills & upwards :crazy: ) and larger different mix ferrite thicker silver plated teflon wire a low loss solid groundplane to cut as much volt drop as practical, homebrew tuning capacitors,

it worked, even a 2pill with fixed interger turns ratios would whoop a conventional 4x2879, taking on anybody in the same pill class that was using conventional design amps was no problem,
i have keyed my 2x2879 on just 18v against hot rodded kl500's on stock volts many times and never lost a keydown,

keydown style amps are far from optimal if all you have is stock voltage, they are very inefficient on stock voltage sucking more current per watt output than a conventional amp with double the number of 2879's, been class C they also sound like **** on SSB and rough on AM,

the copycats like xfarce & fatboy sell you the same amp for stock voltage or keydown use, they also put SSB delays on class C amps which is stupid & proves they don't understand much about how amplifiers work.

many folk think that because their amp is class C it must be more efficient than a class AB amp,
if both are correctly designed for the same supply voltage that's true, but its far from true when the class C amp is designed for 21v+ and running on stock voltage,

an interesting read for the folk who don't believe it.
https://forums.qrz.com/index.php?thread ... ha.248732/

User avatar
sureshot
Veteran
Veteran
Posts: 2675
Joined: 24 Jan 2012, 21:26
Call Sign: 26TM413
Location: South East Coast UK.

Re: US amplifiers.

Post by sureshot » 03 Jun 2018, 10:10

wa10 wrote:
02 Jun 2018, 21:29
"Few of the American amplifier builder know anything about RF design"

absolutely true 148,

many just copy davemade amps ( swing boxes ) that never were designed for cleanliness or running on stock voltage, the design is nothing like any conventionally designed amp from helge grandbergs motorola application notes or the toshiba datasheet,

the amps were designed to win keydowns against conventional amps which they will do, a 2pill dave style amp WILL whoop 4pill in a keydown,

datasheet nuthuggers are armed with a little info but not enough, toshiba datasheets only tell half the story, look at other transistors that show output & imd at different supply voltages,

Dave is smart, he understood vcc vs load impedance years ago as anybody that builds tube amps should understand, he knew doubling your supply voltage quadruples your output power when optimally load matched so a moderate increase in vcc can yield much more power,

as you raise vcc optimum load impedance increases, change class & optimal load impedance changes,

dave knew that you can run transistors at higher vcc and load impedance at a reduced duty cycle to get more power without exceeding maximum device current or maximum device dissipation, how much more power ? a hell of a lot more than the datasheet tells us,

that its not just current but current x time that smokes the device,
a fuse will handle many times its amp rating if you pulse current through it so reducing the time the device is turned by going class C allows the gold wires that connect the tabs to the die under the beryllium oxide cap that coined the term pills to handle more current than the datasheet tells us,

you can push a device into deep class C by using negative voltage biasing to reduce on time even further than a grounded base type class C,
a few 55 built amps used a pp3 battery hidden inside the amp to get the negative biasing,

Dave understood what the datasheet did and more importantly what it did not tell you, and that absolute ratings are higher than what the datasheet tells us if reliability and imd are of no concern,

you don't build the same engine for a drag race if you want to race round a track for many laps and a different engine again if you want to go shopping or are the slightest bit concerned about emissions, there's no over fuelling to cool cylinder temps in your shopping trolly,

he built class C drag racing amps with the required for keydowns 2sc2879's that ran on 21v or a tad higher ( 18v max on datasheet ) with transformer turns ratios and tuning ( oh the top secret tuning on 6pills & upwards :crazy: ) and larger different mix ferrite thicker silver plated teflon wire a low loss solid groundplane to cut as much volt drop as practical, homebrew tuning capacitors,

it worked, even a 2pill with fixed interger turns ratios would whoop a conventional 4x2879, taking on anybody in the same pill class that was using conventional design amps was no problem,
i have keyed my 2x2879 on just 18v against hot rodded kl500's on stock volts many times and never lost a keydown,

keydown style amps are far from optimal if all you have is stock voltage, they are very inefficient on stock voltage sucking more current per watt output than a conventional amp with double the number of 2879's, been class C they also sound like **** on SSB and rough on AM,

the copycats like xfarce & fatboy sell you the same amp for stock voltage or keydown use, they also put SSB delays on class C amps which is stupid & proves they don't understand much about how amplifiers work.

many folk think that because their amp is class C it must be more efficient than a class AB amp,
if both are correctly designed for the same supply voltage that's true, but its far from true when the class C amp is designed for 21v+ and running on stock voltage,

an interesting read for the folk who don't believe it.
https://forums.qrz.com/index.php?thread ... ha.248732/
I didn't realise it would be that complicated, well it looks a lot more in depth than the tutorials i watched. I've never heard or seen any amplifier being driven much above 14 Volts. That's not to say its not possible, RF as a topic is not my strongest point. So i agree, i think some more reading and literature might be needed first. I'm intrested in mosfet driven amps also as much as bjt's. AM is of limited use to me, so any amps built for that not worth doing. Its FM and SSB that's the mainstay in the UK. Thanks for the in depth explanation, i certainty need to find out more in literature.
CB call. Shipwreck.

wa10
Top Poster
Top Poster
Posts: 1169
Joined: 19 Feb 2008, 00:37

Re: US amplifiers.

Post by wa10 » 03 Jun 2018, 19:53

sureshot,

building amplifiers seems like a good idea & you get the satisfaction of "i built that" ,

i don't know what style amps you have been looking at building, i just had a quick look at one instructional video on youtube using 1xsd1446 to drive 2x2879's,
sd1446 is a poor choice for HF its a vhf transistor with lots of gain on HF,

even RM hla amps using 1446's are not unconditionally stable, they can oscillate on certain frequencies with moderate not crazy loads, gain rises significantly as you go down in frequency because they use no gain compensating feedback networks, they are cb amps with filters,

the video is the same style of construction as the keydown amps minus the large ferrite transformers & semco metal capacitors,
they are not linear amps,
a properly designed 2x2879 class AB amp does not need the driver stage, 2879's in a proper circuit have plenty of gain for been driven by a cb radio to sensible output levels,

2pill keydown amps on 14V only manage about 10db gain, its class C so sounds like ass on ssb.

User avatar
sureshot
Veteran
Veteran
Posts: 2675
Joined: 24 Jan 2012, 21:26
Call Sign: 26TM413
Location: South East Coast UK.

Re: US amplifiers.

Post by sureshot » 03 Jun 2018, 22:09

Yes it seems it's these C class amps I've been watching tutorials on. Not best suited to FM or SSB by the looks of it. Shame as they look reasonably straight forward to build, given the correct specifications. But it looks like they are mainly used on AM, I'd missed that bit along the way. Totally forgot they don't do much FM in the US. I'm certain I don't have enough to design one from scratch, I'm going to have a think on it. I know, I think I got caught up in the wow factor of the these 2879 monster amplifiers. Oh well it was fun while it lasted. I might do better to find a schematic of a production amplifier, see if I can go over it and work through each stage. I might be in a better position to put something together then.
CB call. Shipwreck.

User avatar
Unit 148 Mobile
Super Member
Super Member
Posts: 110
Joined: 24 Sep 2010, 20:50
Location: London
Contact:

Re: US amplifiers.

Post by Unit 148 Mobile » 16 Jun 2018, 15:57

Sureshot,

I found something that should suit you and give you an intro into building RF linear amplifiers, and it comes biased suitable for SSB operation should you wish.

Keep an eye for this guys offerings on eBay or contact him direct with any questions you may have: http://www.telstar-electronics.com/index.html

You can read his posts on his FB page which makes for interesting reading: https://www.facebook.com/Telstar-Electr ... 441516710/

Unit 148

User avatar
sureshot
Veteran
Veteran
Posts: 2675
Joined: 24 Jan 2012, 21:26
Call Sign: 26TM413
Location: South East Coast UK.

Re: US amplifiers.

Post by sureshot » 17 Jun 2018, 22:44

Unit 148 Mobile wrote:
16 Jun 2018, 15:57
Sureshot,

I found something that should suit you and give you an intro into building RF linear amplifiers, and it comes biased suitable for SSB operation should you wish.

Keep an eye for this guys offerings on eBay or contact him direct with any questions you may have: http://www.telstar-electronics.com/index.html

You can read his posts on his FB page which makes for interesting reading: https://www.facebook.com/Telstar-Electr ... 441516710/

Unit 148
Thanks for the links, I will certainly give it a look. I'm a noob at this, so any help greatly appreciated. :thumbup:
CB call. Shipwreck.

Kaliphan
Regular
Regular
Posts: 54
Joined: 26 Aug 2016, 09:35
Call Sign: M0

Re: US amplifiers.

Post by Kaliphan » 18 Jun 2018, 12:20

Be wary of anyone who doesn't provide a schematic* though, unless you get different instructions with the kit there's nothing in the construction manual and Telstar don't even specify the part numbers of the FETs in the manual.

IMHO you might be better off with one of the Chinese kits from eBay which would seem to be pretty much the same thing and a heck of a lot cheaper.

*You can, with some effort, reverse the schematic from the manual if you wish but from the bit I drew out,it's not that different from this one:

http://pa-11019.blogspot.com/2016/11/di ... power.html

User avatar
sureshot
Veteran
Veteran
Posts: 2675
Joined: 24 Jan 2012, 21:26
Call Sign: 26TM413
Location: South East Coast UK.

Re: US amplifiers.

Post by sureshot » 20 Jun 2018, 22:36

Kaliphan wrote:
18 Jun 2018, 12:20
Be wary of anyone who doesn't provide a schematic* though, unless you get different instructions with the kit there's nothing in the construction manual and Telstar don't even specify the part numbers of the FETs in the manual.

IMHO you might be better off with one of the Chinese kits from eBay which would seem to be pretty much the same thing and a heck of a lot cheaper.

*You can, with some effort, reverse the schematic from the manual if you wish but from the bit I drew out,it's not that different from this one:

http://pa-11019.blogspot.com/2016/11/di ... power.html
I like the look of that, it looks something just right for a newbie like me :D Thanks for posting that up.
Worst case scenario, build an amplifier from unknown schematic and it doesn't work, or has problems.
CB call. Shipwreck.

User avatar
Unit 148 Mobile
Super Member
Super Member
Posts: 110
Joined: 24 Sep 2010, 20:50
Location: London
Contact:

Re: US amplifiers.

Post by Unit 148 Mobile » 11 Jul 2018, 19:57

Kaliphan,

I think you must be thinking of another Telstar Electronics.

Not only does he state which devices he uses in ALL of his kits but he emphasizes he only develops kits where the output transistors have a datasheet from a reputable manufacturer.

Here is an example of an 'unrecognized' manufacturers ripping off another device: https://www.rfparts.com/2sc2879-rfp.html

The only recognized manufacturer of the 2SC2879 is Toshiba and they no longer produce this part. Here is the 'official' datasheet: http://www.datasheetcatalog.com/datashe ... 2879.shtml

Unit 148

Kaliphan
Regular
Regular
Posts: 54
Joined: 26 Aug 2016, 09:35
Call Sign: M0

Re: US amplifiers.

Post by Kaliphan » 12 Jul 2018, 14:58

Nope, I followed the link you provided and the Skywave DX150 is pretty much the same as the Chinese kit I linked.

The DX250 is a better product.

User avatar
Unit 148 Mobile
Super Member
Super Member
Posts: 110
Joined: 24 Sep 2010, 20:50
Location: London
Contact:

Re: US amplifiers.

Post by Unit 148 Mobile » 14 Jul 2018, 06:20

Kaliphan,

You're sidestepping the point.

The point being made by me was that Telstar Electronics uses devices from reputable semiconductor manufacturers and not 'no-name' manufacturers or rebranders!

Regarding the comparison the the eBay advertised Chinese 70W amp and that of the Telstar DX150.

They are completely different animals as one can clearly see.

The output devices are different + the Chinese amplifier claims of being a broadband design whereas the Telstar amplifier, as with all of his amplifiers are optimized (with correct impedance matching to offer a close to perfect match) for use in the 27MHz CB band.

This is but two of the many differences in the two amplifiers.

You may wish to look again at the two schematics.

Unit 148

Kaliphan
Regular
Regular
Posts: 54
Joined: 26 Aug 2016, 09:35
Call Sign: M0

Re: US amplifiers.

Post by Kaliphan » 14 Jul 2018, 07:28

Unit 148 Mobile wrote:
14 Jul 2018, 06:20
Kaliphan,

You're sidestepping the point.

The point being made by me was that Telstar Electronics uses devices from reputable semiconductor manufacturers and not 'no-name' manufacturers or rebranders!

Regarding the comparison the the eBay advertised Chinese 70W amp and that of the Telstar DX150.

They are completely different animals as one can clearly see.

The output devices are different + the Chinese amplifier claims of being a broadband design whereas the Telstar amplifier, as with all of his amplifiers are optimized (with correct impedance matching to offer a close to perfect match) for use in the 27MHz CB band.

This is but two of the many differences in the two amplifiers.

You may wish to look again at the two schematics.

Unit 148
I would, if he provided one but he doesn't, which was my point, I derived the schematic from the construction manual.

I would not spend money or recommend spending money on a kit before I'd seen the schematic, if you don't get to see it then you can't make any kind of judgement.

The Chinese amps at least show you the schematic first compete with named parts and they look to have decent impedance matching too.

The Telstar amp is a different board but I can't, from reverse engineering of the schematic out of the construction details, say it varies in any significant detail from the *far* cheaper Chinese version, it looks almost identical.

Oh, and the Huagao transistors?

The American 'pill boys' rate them.

User avatar
Unit 148 Mobile
Super Member
Super Member
Posts: 110
Joined: 24 Sep 2010, 20:50
Location: London
Contact:

Re: US amplifiers.

Post by Unit 148 Mobile » 16 Jul 2018, 03:44

Kaliphan,

Telstar has in the past published his schematics on his Facebook page page but for good reason, and you can imagine why he has decided to delete them after a short period. I would do the same otherwise how would one be compensated from all the design time?

Regarding the differences between the eBay Chinese 70W amplifier and the Telstar. I have already listed two differences above but did not wish to use excess bandwidth to make my point. With your component level design experience you can see the differences with your own eyes. These are two very basic amplifiers. The only thing I cannot deduce from the two images are of course component values and turns ratio of the input and output transformers but this can be reverse engineered from the semi's datasheet and knowing you require a 50 ohm input & output impedance. As you'd know though with RF mathematics is only 90% of the equation. Final tweaking of the design is the other 10%.

As far as operation across the 27MHz band I'm willing to put my money where my mouth is and bet that Telstars amplifier comfortably outperforms the Chinese one both of power in for power out (at the rated RF input power and operating voltage). Shall we say £100.00 GBP ($130.00 at today's rates) should you like a wager. My neighbor Adrian is a like myself a licensed amateur and he is a retired chief engineer at the UK's GCHQ. He has my Bird 4410 watt-meter and portable HP spectrum analyser. I'd be very happy to ask him to carry out an independent test with a video for publication here with the mod's permission and verification if you're willing to step up to the mark. I will purchase both amplifiers out of my pocket. I am confident, in fact I'm more than confident of the winning amplifier.

May I suggest the loser sends their £100.00 to a children's charity of their choice?

Unit 148 Mobile

Post Reply

Who is online

Users browsing this forum: No registered users and 5 guests