Speedy. The variable speed accellerator.

DDNI · 04 June 2012, 10:53

Has anyone ever tried one of these variable speed oscillators on an amiga accellerator?

http://www.micromac.com/products/speedy.html

roy bates · 04 June 2012, 11:53

i was looking for one of these a few months back,does anyone know the frequency range of one of these?

it would have to work(adjustable) between 50-120mhz in my case.

and what are the incraments between frequencys i wonder.

DDNI · 04 June 2012, 11:59

Looking at the table on the link, 30 - 125mhz is the range. It also seems to be fine enough to deliver speeds which involve single unit changes.

Availability is my big question. The link is old...

roy bates · 04 June 2012, 12:03

yeah thats what im worried about,the link is pretty old.

it should work as it replaces the can oscillator,and it is 5 volt and ttl by the look of it.

Stedy · 04 June 2012, 12:42

One of these devices from Linear Technology would work,
http://www.linear.com/product/LTC6905
and you can purchase it from Farnell.

A small PCB and it could fit 8/14 pin DIL sockets.

Ian

roy bates · 04 June 2012, 13:00

looks interesting i might give one of these a go stedy.

Stedy · 06 June 2012, 22:59

Hi,

Just checked the workshop stores, I have a couple of LTC6907 (40KHz to 4 MHz), bought for another project I will test shortly. The 1-2% accuracy does concern me a little but for a CPU it should be OK. I typically use around 0.0035% accuracy crystals.

Will report back with test results.

Ian

roy bates · 06 June 2012, 23:31

hi stedy,do you think it would be better to run a LTC6905 on 3.3 volts rather than 5 volts.
i dont know how it will fair between 50 and 110mhz.

Stedy · 18 June 2012, 20:53

Hi,

I tested the LTC6907 device I had and was pleased with the results. Initial frequency accuracy was within tolerance (0.7% compared to a potential 1%) and waveform shape/symmetry was good. Would like to try it with some 0.1% resistors, when I buy some.

@roy bates
The LTC6905 can use 3.3V or 5V, which is great. The 68060 is 3.3V and I would expect the clock source to also be 3.3V LVTTL. My only concern with using 3.3V with a 5V part, is if the CPU or CPLD has HCMOS inputs, so requires 3.33V minimum (66% of 5V) for a logic 1. Would need to check all potential devices.

Using the LTC6905 is easy enough though. Could easily make some adaptors, if there was the demand.

Ian

roy bates · 18 June 2012, 21:33

it shouldent be a problem really, all 040 and 060 cards oscilators on amigas are 5 volts ,i was just wondering if it could be run on 3.33 volts and still be usable.

04 June 2012, 10:53	#1
DDNI Targ Explorer Join Date: Mar 2006 Location: Northern Ireland Posts: 5,433	Speedy. The variable speed accellerator. Has anyone ever tried one of these variable speed oscillators on an amiga accellerator? http://www.micromac.com/products/speedy.html

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Join variable and string in scripts.	olesio	support.Apps	3	31 January 2013 11:44
NTSC variable-length scanlines	TheDarkCoder	Coders. Asm / Hardware	3	23 November 2011 15:51
Shoot the Speedy One...	JonSick	request.Old Rare Games	0	16 September 2010 12:14
What accellerator do I need ?	Kakaboy	Hardware mods	13	23 March 2010 04:33
Variable Master Clock	wiltshireguyuk	request.UAE Wishlist	0	13 December 2004 16:40

04 June 2012, 11:53	#2
roy bates Registered User Join Date: Apr 2011 Location: birmingham Age: 55 Posts: 2,827	i was looking for one of these a few months back,does anyone know the frequency range of one of these? it would have to work(adjustable) between 50-120mhz in my case. and what are the incraments between frequencys i wonder.

04 June 2012, 11:59	#3
DDNI Targ Explorer Join Date: Mar 2006 Location: Northern Ireland Posts: 5,433	Looking at the table on the link, 30 - 125mhz is the range. It also seems to be fine enough to deliver speeds which involve single unit changes. Availability is my big question. The link is old...

04 June 2012, 12:03	#4
roy bates Registered User Join Date: Apr 2011 Location: birmingham Age: 55 Posts: 2,827	yeah thats what im worried about,the link is pretty old. it should work as it replaces the can oscillator,and it is 5 volt and ttl by the look of it.

04 June 2012, 12:42	#5
Stedy Registered User Join Date: Jan 2008 Location: United Kingdom Age: 46 Posts: 733	One of these devices from Linear Technology would work, http://www.linear.com/product/LTC6905 and you can purchase it from Farnell. A small PCB and it could fit 8/14 pin DIL sockets. Ian

04 June 2012, 13:00	#6
roy bates Registered User Join Date: Apr 2011 Location: birmingham Age: 55 Posts: 2,827	looks interesting i might give one of these a go stedy.

06 June 2012, 22:59	#7
Stedy Registered User Join Date: Jan 2008 Location: United Kingdom Age: 46 Posts: 733	Hi, Just checked the workshop stores, I have a couple of LTC6907 (40KHz to 4 MHz), bought for another project I will test shortly. The 1-2% accuracy does concern me a little but for a CPU it should be OK. I typically use around 0.0035% accuracy crystals. Will report back with test results. Ian

06 June 2012, 23:31	#8
roy bates Registered User Join Date: Apr 2011 Location: birmingham Age: 55 Posts: 2,827	hi stedy,do you think it would be better to run a LTC6905 on 3.3 volts rather than 5 volts. i dont know how it will fair between 50 and 110mhz.

18 June 2012, 20:53	#9
Stedy Registered User Join Date: Jan 2008 Location: United Kingdom Age: 46 Posts: 733	Hi, I tested the LTC6907 device I had and was pleased with the results. Initial frequency accuracy was within tolerance (0.7% compared to a potential 1%) and waveform shape/symmetry was good. Would like to try it with some 0.1% resistors, when I buy some. @roy bates The LTC6905 can use 3.3V or 5V, which is great. The 68060 is 3.3V and I would expect the clock source to also be 3.3V LVTTL. My only concern with using 3.3V with a 5V part, is if the CPU or CPLD has HCMOS inputs, so requires 3.33V minimum (66% of 5V) for a logic 1. Would need to check all potential devices. Using the LTC6905 is easy enough though. Could easily make some adaptors, if there was the demand. Ian

18 June 2012, 21:33	#10
roy bates Registered User Join Date: Apr 2011 Location: birmingham Age: 55 Posts: 2,827	it shouldent be a problem really, all 040 and 060 cards oscilators on amigas are 5 volts ,i was just wondering if it could be run on 3.33 volts and still be usable.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)