By continuing to use this site, you agree to our use of cookies. Find out more
Forum sponsored by:
Forum sponsored by Allendale Oct 22nd

Anyone good at fault finding with amplifiers here?

All Topics | Latest Posts

Search for:  in Thread Title in  
Neil Wyatt18/08/2019 23:09:09
avatar
Moderator
16662 forum posts
687 photos
75 articles
Posted by Neil Wyatt on 18/08/2019 23:06:44:
Posted by Mark Rand on 18/08/2019 22:23:51:

But you don't have to repair them. They are so cheap that they are disposable.

Exactly.

The Trace Elliot Elf is the descendant fo my amp, 200W and it fits in your pocket - literally. But not attached to a 4x10 cab of course.

Michael Gilligan18/08/2019 23:50:45
avatar
14150 forum posts
616 photos

Posted by Neil Wyatt on 18/08/2019 23:09:09:

.

The Trace Elliot Elf is the descendant fo my amp, 200W and it fits in your pocket - literally. ...

.

Did they take two of the knobs off to get it in her back pocket ?

**LINK**

http://www.traceelliot.com/

MichaelG.

Joseph Noci 119/08/2019 07:32:49
543 forum posts
832 photos
Posted by Mark Rand on 18/08/2019 22:23:51:

But you don't have to repair them. They are so cheap that they are disposable. I used to be able to hear 20kHz as a teenager, but I could never hear 200kHz...

Well, You are then very lucky in the UK...

Most of the Combo Amps, PA systems and Jam Amps I get to repair here are Peavey, Phonic, and ALTO.

They are considered quite good stuff, and they have shifted most of their classis amps to Class D ( Peavey lagging behind, thank goodness).

A typical 500W+500W amp from Phonic or ALTO cost approx £650.00 . A 'Digital' (class D) version costs around £900.00 . An analogue powered speaker ( Woofer/Sub with a Tweeter) of 400Watts is around £400.00 in 'analogue' and around £600.00 in class D.

When I chase for spares, the agents in South Africa always start there response with 'Ooo, thats a Digital Amp, quite pricey, spares are..', or words to the same effect!

And of course, the next Hate up the list is a class D amp WITH a switchmode powersupply...THAT the agents refer to as a FULLY Digital amp....

And the trend with the agents these days is to not keep component spares - for most of the mid-range power amps they keep a powers supply module and the amplifier module with the latter around 35% the cost of the complete unit..

The only thing to like about that technology is that at my age it's easier to carry up the stairs to the workshop..

And, as a radio ham, I hate them even more..The EMI many of these amps generate on the HF bands is terrible! The filtering is always done down to a cost, with lip service to the regulations..Some of them are dead quiet till you plug in the guitar or a microphone..

Bring back the good Mr Hood anyday!

Rant rant...

Joe

Joseph Noci 119/08/2019 07:51:45
543 forum posts
832 photos

By the way, Neil, Your amp seems to have been a Maplin Kit in the late 1980's - the circuit and some descriptive comments are in Ian Hickman's book 'Analog Circuitry Explained' - published 1990.

Joe

Neil Wyatt19/08/2019 10:02:51
avatar
Moderator
16662 forum posts
687 photos
75 articles
Posted by Joseph Noci 1 on 19/08/2019 07:51:45:

By the way, Neil, Your amp seems to have been a Maplin Kit in the late 1980's - the circuit and some descriptive comments are in Ian Hickman's book 'Analog Circuitry Explained' - published 1990.

Joe

I've got Michael Tooleys' Maplin Electronic Circuits Handbook, which has a section on that kit. I came within a whisker of making it a few times!

The basic idea is the same, although the biaising is different and there are detail component differences everywhere. There's also a design for a rather less powerful (about 40W RMS) version using a pair of power darlingtons earlier in the book and between the two descriptions you can get a good idea how this configuration works. Essentially, they (and the Trace Elliot design) are high-power op amps made with discrete components! Relative simplicity, good zero crossing performance and accurate feedback makes them very low THD and excellent for demanding applications.

I've made two of the bipolar version, somewhere I have a design for the board, one to replace the power amp in a Carlsboro guitar amplifier and one for a little 'uAmp Micro Bass Combo' I made from scratch.

Back to the TE design, mine is the original 150W version with two fets and a fan. The ones I have the schematics for include a 150W version with four FETS and no fan and a 250W one with four FETs and a fan.

I could uprate to 250W simply by getting hold of an uprated transformer and two FETS and two resistors...

Neil

AdrianR19/08/2019 17:01:01
272 forum posts
20 photos

The Maplin amp from the 80's, sounds like the ones I built. Fantastic little amps. With the preamps I designed I had so little hum. Even at full volume and no input you had to put your ear against the speaker to hear it.

Now a 4Ohm 100W resistor for a dummy load has turned up, my next job is to fix a 100W Traynor valve amp. Will be fun to play with valves again.

Adrian

Neil Wyatt20/08/2019 00:03:26
avatar
Moderator
16662 forum posts
687 photos
75 articles

The +/- 69V DC power rails seem awfully generous for a 150W amp.

I just had a look, and the amp is powered by a massive transfroer from the 50-0-50VAC terminals, not the 39-039V ones that would suit a 150W amp.

I'm now wondering if the reason it wasn't fitted with sockets for an external cab was because it only has two instead fo 4 mosfets.

Strikes me it would be sensible to fit a second pair seeing as the PCB has the locations for them if I'm running it at 4R.

Neil

Neil Wyatt20/08/2019 00:37:13
avatar
Moderator
16662 forum posts
687 photos
75 articles

I found a thread, in Dutch, about someone with the same problem, except their DC offset was only 127mV.

Much discussion and debate until in the last post they replaced TR201/2 with a matched pair and the offset came down to 5mV, then replaced TR203/4 as well and it came down to 3mV.

I have high hopes...

There should be a translated version of the thread here:

Thread on Dutch forum

Neil

Joseph Noci 120/08/2019 14:54:29
543 forum posts
832 photos

I see that Dutch fellow did sort of as I asked you to do - gently heat the long tailed pair individually with a soldering iron - except he cooled them with a frozen pancake ....Ah well, whatever flaps your jack...

I am sure your problem is simply gain mismatches in the long tailed pair(s).

The math shows it, so why not. Just fiddled a bit with LTP's in LTSpice ( Linear Technologies, now Analog Devices) free version of Spice - just cause its easy, lazy and quick.. Playing with the gains, with large loop gains, give BIG offsets for small hfe deltas.

When will you get your tranny's?

What intrigues me is ( assuming the amp's original offset was 'good' ) what causes it to change? Unless the transistor is abused, over-voltaged, overheated, etc, hfe should not really change much in years, and in a matched pair, the changes should be similar in both. From the circuit values I've seen , typical of your amp, current levels and Vce are well under control, so no abuse there. Interesting. I also remember, many years ago when making accessory widgets for our 'band' , I made a white noise generator, feeding voltage controlled filters to generate ocean sounds..but, noise eluded the device, Thinking the device stuffed, I used the transistor tester to verify - The transistor was ok, with hfe around 370. So I took a cigarette lighter and heated the BC107 till it glowed .. after cooling, I tested it again, still ok..and hfe had changed to 368....!!..within measurement error...

But,boy, did it now make nice ocean sounds....

Joe

SillyOldDuffer20/08/2019 15:14:46
4789 forum posts
1011 photos

Posted by Joseph Noci 1 on 20/08/2019 14:54:29:

...

What intrigues me is ( assuming the amp's original offset was 'good' ) what causes it to change? Unless the transistor is abused, over-voltaged, overheated, etc, hfe should not really change much in years, and in a matched pair, the changes should be similar in both.

...

But,boy, did it now make nice ocean sounds....

Joe

Maybe the can's seal gradually failed over the years and the innards are exposed to dirt and moisture. Perhaps Neil's transistor has gone rusty.

Ocean sounds invariably produced by my amplifier efforts too. Motorboating...

Dave

Neil Wyatt20/08/2019 17:38:16
avatar
Moderator
16662 forum posts
687 photos
75 articles

You can tell it was a Dutch forum. A Frenchman woudl have used a frozen crepe, and a Belgian would have pointed a hot waffle at it.

Hopefully replacements arrive tomorrow.

It seems other folks have added the 'missing' power FETs as well.

I remember removing the cans from AC127s to make phototransistors - choose the ones where the blue heatsink compound doesn't cover all the chip.

Neil

Neil Wyatt20/08/2019 18:11:21
avatar
Moderator
16662 forum posts
687 photos
75 articles

Went back through some old threads where I had been going on about how the amp was putting out a lot more power than its rated 150W (the legendary 'Trace Elliot Watts'.

Found a post by an ex-TE engineer I should have read properly.

Some 150W two-fet boards were put out with the PSUs for 200W/300W into 4ohm amps.

With a 50-0-50 transformer and PSU capacitors rated 4700uf/100V instead of 63V mine is obviously one of these 'snowflake' amps.

Presumably this is why the external speaker sockets weren't fitted but a fan was.

So add my two extra FETs and I have a rare-as-hen's teeth TE 300W combo.

Neil

AdrianR20/08/2019 18:37:36
272 forum posts
20 photos

Back in the day when making differential amps it was common to glue the two transistors together so the temperatures were the same. I doubt your board will allow that, but you could glue a small bit of aluminium or copper between the pairs.

If you still have an offset you could inject a DC offset to T202. As you dont have a separate regulated supply, you will need to decouple it with a capacitor.

Adrian

Neil Wyatt20/08/2019 18:42:04
avatar
Moderator
16662 forum posts
687 photos
75 articles
Posted by AdrianR on 20/08/2019 18:37:36:

Back in the day when making differential amps it was common to glue the two transistors together so the temperatures were the same. I doubt your board will allow that, but you could glue a small bit of aluminium or copper between the pairs.

If you still have an offset you could inject a DC offset to T202. As you dont have a separate regulated supply, you will need to decouple it with a capacitor.

Adrian

I'll be happy to get it to a fraction of a volt, it's not hifi, I just want to make sure i can sue all the headroom and keep it running a bit cooler

Neil

Neil Wyatt21/08/2019 14:39:28
avatar
Moderator
16662 forum posts
687 photos
75 articles

The replacement transistors have come.

The gain spec for the BF423 is >50, the two fitted are 80 and 64, but is suspect at least one is partially fried.

Out of interest the new ones are (first measurements, repeats come out +/-1):

129 - 2

134 - 2

135 - 2

136 - 1

138 - 1

139 - 1

151 - 1

The 151 is interesting, otherwise they are all pretty close for a random selection.

And... the output offset voltage is now 0.22V, a bit higher than I hoped, but now no visible displacement of the speaker cones. I will replace the pair of BF422s and adjust the bias current when I fit the two extra FETS, but I need to wait for the mounting kits to arrive.

Neil

Joseph Noci 121/08/2019 15:06:59
543 forum posts
832 photos

That's a good result Neil. So, You found the mismatch and there was not much hfe from the rest of us...

Joe

 

- if you want to get the offset down, and mind not being a scoundrel, warm each transistor with a soldering iron ( did I say this before??) - one of them will make the offset reduce - then, the scoundrel part, heat that one hot and let it cool and see where the offset ends up - repeat with greater heat each time till happy..

- You mentioned the phototransistor trick - I did the same but used OC70's - nicked out of modern transistor/valve hybrid car radios, themselves nicked from the local scrapyard where cars went to die. Remember those mechanical vibrators for the valve HT? And OC35's in a multivib stage that replaced the vibrators? Got a lot of electronic bits from that scrapyard's car radios in my kid days..

Edited By Joseph Noci 1 on 21/08/2019 15:16:34

Neil Wyatt21/08/2019 16:02:08
avatar
Moderator
16662 forum posts
687 photos
75 articles

I heard of those HT units, but don't recall seeing one.

I bought a (new) dual valve preamp a few months ago, just because it was ridiculously cheap, runs off 12V stepped up with a more modern charge pump on board. Ideal for sticking a valve overdrive emulatoir into a transistor guitar amp, methinks. Only has a gain of about 2, apparently (no documentation with it), but easy enough to stick a high-gain transistor stage in front.

"so many projects, so little time"...

Bandersnatch21/08/2019 17:28:33
avatar
1267 forum posts
40 photos
Posted by Neil Wyatt on 21/08/2019 16:02:08:

"so many projects, so little time"...

.... wait till you get to the average age around here wink

[ "Age" might make an interesting (anonymous) poll. ]

Joseph Noci 122/08/2019 07:09:08
543 forum posts
832 photos
Posted by Neil Wyatt on 20/08/2019 18:11:21:

Went back through some old threads where I had been going on about how the amp was putting out a lot more power than its rated 150W (the legendary 'Trace Elliot Watts'.

Found a post by an ex-TE engineer I should have read properly.

Some 150W two-fet boards were put out with the PSUs for 200W/300W into 4ohm amps.

With a 50-0-50 transformer and PSU capacitors rated 4700uf/100V instead of 63V mine is obviously one of these 'snowflake' amps.

Presumably this is why the external speaker sockets weren't fitted but a fan was.

So add my two extra FETs and I have a rare-as-hen's teeth TE 300W combo.

Neil

I presume your posted schematic is complete in its output stage, ie, not a 'representation'? I would expect good design would have placed current sharing resistors in each FET output leg when more FETs are added in parallel.

Not sure it is a good idea omitting these resistors..(0.2ohm to 0.33ohm 5watt for your amp)

Joe

AdrianR22/08/2019 08:06:20
272 forum posts
20 photos

I presume your posted schematic is complete in its output stage, ie, not a 'representation'? I would expect good design would have placed current sharing resistors in each FET output leg when more FETs are added in parallel.

Not sure it is a good idea omitting these resistors..(0.2ohm to 0.33ohm 5watt for your amp)

Joe

Joe,

It is vital to have the resistors for BJTs, as they have a positive temperature coefficient which means the current increases with temperature. MOSFETs do not need the resistors to compensate for temperature effects. MOSFETs have a negative temperature coefficient which means the current decreases as they get warm, this means they self balance.

Adrian

All Topics | Latest Posts

Please login to post a reply.

Magazine Locator

Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!

Find Model Engineer & Model Engineers' Workshop

Support Our Partners
Ausee.com.au
Warco
cowbells
TRANSWAVE Converters
Eccentric Engineering
Meridienne; London MES
ChesterUK
Eccentric July 5 2018
emcomachinetools
Allendale Electronics
Subscription Offer

Latest "For Sale" Ads
Latest "Wanted" Ads
Get In Touch!

Do you want to contact the Model Engineer and Model Engineers' Workshop team?

You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.

Click THIS LINK for full contact details.

For subscription issues please see THIS LINK.

Digital Back Issues

Social Media online

'Like' us on Facebook
Follow us on Facebook

Follow us on Twitter
 Twitter Logo

Pin us on Pinterest